Communication method, apparatus and system

Abstract

The present application provides a communication method, apparatus and system. A network device can associate a first resource configuration, P second resource configurations, and a third resource configuration with one another by means of a reporting configuration, thereby enabling a terminal device to determine, on the basis of the reporting configuration, inference results for reference signal resource sets configured for the plurality of resource configurations. Therefore, the network device does not need to send the reporting configuration separately for each resource configuration, thereby reducing configuration overhead.

Description

Communication methods, devices and systems

[0001] This application claims priority to Chinese Patent Application No. 202510395233.4, filed with the China National Intellectual Property Administration on March 28, 2025, entitled "Communication Method, Apparatus and System", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of communications, and more particularly to a communication method, apparatus and system. Background Technology

[0003] Currently, artificial intelligence (AI) has been introduced into wireless communication networks and is widely used in many application scenarios of air interface technology, such as channel state information (CSI) feedback, CSI prediction, beam management, and positioning. When applying AI in beam management, terminal devices or network devices can use AI models to predict information about all beams (e.g., beams in set B) based on measurements of a subset of beams (e.g., beams in set B). For example, it can output the identifiers of the top K beams in set A based on their quality ranking. This can significantly reduce the overhead of beam measurement.

[0004] However, in some scenarios, such as interference-cooperation scenarios, network devices may only use a subset of beams from Set A for transmission within a specific cell. These devices prefer to obtain information from only a portion of Set A (i.e., a subset of Set A) and are less concerned with the overall beam performance of Set A. In other scenarios, such as inter-cell interference-cooperation scenarios, network devices may want to obtain both the beam information of Set A and the beam information from a subset of Set A. Therefore, network devices need to use different reporting configurations to instruct terminal devices to report separately for Set A and its subset. This can lead to significant configuration overhead. Summary of the Invention

[0005] This application provides a communication method, apparatus, and system to reduce configuration overhead.

[0006] Firstly, a communication method is provided. This method can be applied to a first device, which may be a terminal device or a component deployed within the terminal device, such as circuits or chips inside the terminal device (e.g., modem chips, also known as baseband chips, or system-on-chip (SoC) chips or system-in-package (SIP) chips containing modem cores, or chips for AI or SoC or SIP chips containing chips for AI, etc.); or, it can be applied to logic modules or software capable of implementing all or part of the functions of the communication device, etc. For ease of understanding and explanation, the following description uses a terminal device as an example of the first device to illustrate the method.

[0007] For example, the method includes: receiving a reported configuration associated with multiple resource configurations, the multiple resource configurations including: a first resource configuration, P second resource configurations, and a third resource configuration, each of the multiple resource configurations being configured to configure a set of reference signal resources, the P second resource configurations and the first resource configuration satisfying that: the beam set corresponding to the reference signal resource set configured by each of the P second resource configurations is a subset of the beam set corresponding to the reference signal resource set configured by the first resource configuration; P is a positive integer; determining first inference information according to the reported configuration, the first inference information indicating the inference result for the reference signal resource sets configured for the first resource configuration and the P second resource configurations respectively, the inference result being obtained based on measurement results, the measurement results being obtained based on measurements of the reference signal resource set configured by the third resource configuration.

[0008] Based on the above scheme, network devices can associate multiple resource configurations through a single reported configuration, thereby facilitating terminal devices to determine the inference result of the reference signal resource set configured for multiple resource configurations based on the reported configuration. Therefore, network devices do not need to send a separate reported configuration for each resource configuration, reducing configuration overhead.

[0009] Optionally, the set of reference signal resources configured in each of the P second resource configurations is a subset of the set of reference signal resources configured in the first resource configuration.

[0010] Optionally, the reference signal resource set configured by at least one of the P second resource configurations is not a subset of the reference signal resource set configured by the first resource configuration, but corresponds to the same beam set.

[0011] It should be understood that the inference result for the first resource configuration may refer to the top beam information inferred from the beam set corresponding to the reference signal resource set configured in the first resource configuration (i.e., the first beam set mentioned below). The inference result for the second resource configuration may refer to the top beam information inferred from the beam set corresponding to the reference signal resource set configured in the second resource configuration (i.e., the second beam set mentioned below).

[0012] In one possible design, the first inference information includes information about multiple beams, which is the complete set of the following terms: K beams in the first beam set, and K beams in the p-th second beam set obtained by iterating through p from 1 to P. p There are p-th beams, the first beam set corresponds to the reference signal resource set configured in the first resource configuration, and the p-th second beam set corresponds to the reference signal resource set configured in the p-th second resource configuration among P second resource configurations, where p, K, and K are the reference signal resource sets configured in the first resource configuration. p It is a positive integer.

[0013] Here, the complete set can refer to the set obtained after deduplicating identical beams. Deduplication reduces the overhead of indicating duplicate beams.

[0014] Optionally, the reported configuration may also indicate one or more of the following: K, K1 to K P .

[0015] Optionally, K, K1 to K P One or more of them are predefined.

[0016] K', K1 to K' are specified through network device indications and / or predefined methods. P Based on the values ​​of each item, the terminal device can determine the number of top beams that need to be indicated for each beam set in the first inference information.

[0017] In another possible design, the first inference information includes: a first inference result and P second inference results; wherein, the p-th second inference result among the P second inference results includes K from the p-th second beam set in the P second beam set. p The information of each beam, wherein the p-th second beam set in the P second beam sets corresponds to the reference signal resource set configured in the p-th second resource configuration in the P second resource configurations; the first inference result includes the information of K' beams in the first beam set, wherein the K' beams are the K' beams in the first beam set excluding the K” beams, and the K” beams are the complete set of the P second beam sets, wherein the first beam set corresponds to the reference signal resource set configured in the first resource configuration; K', K”, K pBoth p and p are positive integers, with p taking values ​​from 1 to p.

[0018] Optionally, the reported configuration may also indicate one or more of the following: K', K1 to K P .

[0019] Optionally, K', K1 to K P One or more of them are predefined.

[0020] K', K1 to K' are specified through network device indications and / or predefined methods. P Based on the values ​​of each item, the terminal device can determine the number of top beams that need to be indicated for each beam set in the first inference information. Since the top beams indicated for the first beam set are determined within the entire set excluding P second beam sets, the overhead of indicating duplicate beams can be avoided. Therefore, more useful beam information can be reported with the same reporting overhead. Furthermore, since the number of top beams determined for each beam set can be predetermined, if the terminal device sends the first inference information to the network device, the network device can also determine the number of top beams determined for each beam set by the first inference information, thereby correctly parsing the first inference information.

[0021] In another possible design, the first inference information includes: a first inference result and P second inference results; wherein, the first inference information includes K from the first beam set. sum Information about each beam; the first inference result includes the information of the beams in the first beam set excluding the K” beams (K). sum -K”) beam information, the first beam set corresponds to the reference signal resource set configured in the first resource configuration; the K” beams are the K beams in the p-th second beam set obtained by traversing p from 1 to P respectively. p The complete set of P beams, the p-th second inference result among the P second inference results includes K in the p-th second beam set of the P second beam sets. p Information about each beam, the p-th second beam set in the P second beam sets corresponds to the reference signal resource set configured in the p-th second resource configuration in the P second resource configurations, K p Greater than or equal to K p,min K p,min Less than K sum ,K”、K sum K p K p,min Both p and p are positive integers, with p taking values ​​from 1 to p.

[0022] Optionally, the reported configuration may also indicate one or more of the following: K sum and K 1,minTo K P,min .

[0023] Optionally, K sum and K 1,min To K P,min One or more of them are predefined.

[0024] K is specified through network device indications and / or predefined methods. sum and K 1,min To K P,min Based on the values ​​of each item, the terminal device can determine the number of top beams that need to be indicated for each beam set in the first inference information. Because K sum Therefore, given a high degree of overlap between the top beams in the first beam set and the top beams in the P second beam sets, the terminal device can indicate as many beams as possible from the second beam sets. Furthermore, if the terminal device sends this first inference information to the network device, the network device can also determine the number of top beams identified by the first inference information for each beam set, thereby correctly parsing the first inference information.

[0025] In conjunction with the first aspect, in some possible implementations of the first aspect, the method further includes: sending first inference information.

[0026] The terminal device can send the first inference information to the network device, thereby enabling the network device to obtain information about the top beams of the first beam set and P second beam sets.

[0027] Optionally, the reporting configuration also indicates a first reporting period, wherein sending the first inference information includes: sending the first inference information at a first reporting time, the first reporting time corresponding to the first reporting period.

[0028] By indicating the first reporting period, terminal devices can easily determine the reporting time for the first inference information. Network devices can also receive the first inference information based on this reporting time.

[0029] In conjunction with the first aspect, in some possible implementations of the first aspect, the reporting configuration further indicates a second reporting period, which is different from the first reporting period, and the first reporting period is an integer multiple of the second reporting period; the method further includes: sending second inference information at a second reporting time, the second reporting time corresponding to the second reporting period, the second inference information including inference results of reference signal resource sets configured for one or more of the P second resource configurations respectively.

[0030] Therefore, terminal devices can report the first inference information and the second inference information based on different reporting periods. Since the second inference information indicates the inference result for the reference signal resource set configured for P second resource configurations, and the first inference information indicates the inference result for the reference signal resource sets configured for the first resource configuration and the P second resource configurations respectively, the network device can, depending on the current scenario, additionally indicate a second reporting period when it needs to obtain the second inference information separately. Alternatively, in scenarios where the network device needs to obtain the inference result for the reference signal resource sets configured for the first resource configuration and the P second resource configurations respectively, the network device can indicate different reporting periods for the first and second inference information. Since the first reporting period is an integer multiple of the second reporting period, the terminal device can report the inference result determined for the P second resource configurations to the network device more frequently, instead of reporting the inference result determined for the first resource configuration every time. Therefore, this satisfies the needs of the network device, ensures performance, and saves signaling overhead.

[0031] Optionally, the reporting configuration also indicates a first reporting period and a second reporting period.

[0032] In conjunction with the first aspect, in some possible implementations of the first aspect, before receiving the reported configuration, the method further includes: sending capability information, the capability information being used to indicate support for determining inference results for a first beam set and a plurality of fourth beam sets, the first beam set corresponding to a reference signal resource set configured by a first resource configuration, each of the plurality of fourth beam sets being a subset of the first beam set, and each of the P second resource configurations having a reference signal resource set configured corresponding to one of the plurality of fourth beam sets.

[0033] The inference result can be obtained by the model performing the first task. Therefore, it can also be said that the capability information is used to instruct the first task to support the determination of inference results for the first beam set and multiple fourth beam sets, or that the capability information is used to instruct the model to support the determination of inference results for the first beam set and multiple fourth beam sets.

[0034] It should be understood that since the reference signal resource set configured in each of the P second resource configurations corresponds to one of the multiple fourth beam sets, that is, the P second beam sets correspond to the P fourth beam sets in the multiple fourth beam sets, it supports determining the inference result for the first beam set and the multiple fourth beam sets, that is, it supports determining the inference result for the first beam set and the P second beam sets, that is, it supports determining the inference result for the first beam set and its subsets.

[0035] By sending capability information, network devices can understand the inference capabilities of the terminal side, and thus configure CSI reports for the terminal devices appropriately.

[0036] In conjunction with the first aspect, in some possible implementations of the first aspect, before receiving the reported configuration, the method further includes: receiving first information, which indicates a first set of reference signal resources, the first set of reference signal resources corresponding to a first beam set; receiving Q pieces of second information, which indicate Q sets of fourth reference signal resources, each of the Q pieces of second information indicating one of the Q sets of fourth reference signal resources, the Q sets of fourth reference signal resources corresponding one-to-one with the Q sets of fourth beam sets, where Q is a positive integer; wherein each of the Q sets of fourth beam sets is a subset of the first beam set, the reference signal resource set configured by one of the P sets of second resource configurations corresponds to one of the Q sets of fourth beam sets, the first beam set corresponds to the reference signal resource set configured by the first resource configuration, and the first set of reference signal resources and the Q sets of fourth reference signal resources are used for training the first task.

[0037] That is, the first reference signal resource set and Q fourth reference signal resource sets are configured using the first information and the second information, respectively, for use in training the first task (or model). It can be seen that the training of this first task considers not only the first reference signal resource set corresponding to the first beam set, but also the fourth reference signal resource sets corresponding to subsets of the first beam set, thus improving the performance of the first task (or model) in predicting subsets of the first beam set.

[0038] In conjunction with the first aspect, in some possible implementations of the first aspect, the method further includes: performing training of the first task based on the first information, the second information, and the third measurement result, wherein the third measurement result is obtained based on measurements of a third set of reference signal resources and includes input data corresponding to the training of the first task.

[0039] Secondly, a communication method is provided, which can be applied to a second device, which can be a network device or a component deployed in the network device, such as circuits or chips inside the network device (e.g., modem chip, or SoC chip or SIP chip containing modem core, or chip for AI or SoC or SIP chip containing chip for AI, etc.); or, it can be applied to logic modules or software that can realize all or part of the functions of the communication device, etc.

[0040] It should be understood that the technical solution of the second aspect corresponds to the technical solution of the first aspect. Therefore, the relevant descriptions and technical effects of the first aspect are equally applicable to the second aspect. Thus, the content that is the same as or corresponding to the first aspect can be referred to the relevant descriptions of the first aspect, and will not be repeated here.

[0041] For example, the method includes: sending a reporting configuration associated with multiple resource configurations, the multiple resource configurations including: a first resource configuration, P second resource configurations and a third resource configuration, each of the multiple resource configurations being used to configure a set of reference signal resources, the P second resource configurations and the first resource configuration satisfying that: the beam set corresponding to the reference signal resource set configured by each of the P second resource configurations is a subset of the beam set corresponding to the reference signal resource set configured by the first resource configuration, where P is a positive integer.

[0042] Optionally, the set of reference signal resources configured in each of the P second resource configurations is a subset of the set of reference signal resources configured in the first resource configuration.

[0043] Optionally, the reference signal resource set configured by at least one of the P second resource configurations is not a subset of the reference signal resource set configured by the first resource configuration, but corresponds to the same beam set.

[0044] In conjunction with the second aspect, in some possible implementations of the second aspect, the method further includes: receiving first inference information indicating inference results for reference signal resources configured for a first resource configuration and P second resource configurations respectively, the inference results being obtained based on measurement results, the measurement results being obtained based on measurements of a set of reference signal resources configured for a third resource configuration.

[0045] Optionally, the reporting configuration also indicates a first reporting period, receiving first inference information, including: receiving first inference information at a first reporting time, the first reporting time corresponding to the first reporting period.

[0046] In one possible design, the first inference information includes information about multiple beams, which is the complete set of the following terms: K beams in the first beam set, and K beams in the p-th second beam set obtained by iterating through p from 1 to P. p There are p-th beams, the first beam set corresponds to the reference signal resource set configured in the first resource configuration, and the p-th second beam set corresponds to the reference signal resource set configured in the p-th second resource configuration among P second resource configurations, where p, K, and K are the reference signal resource sets configured in the first resource configuration. p It is a positive integer.

[0047] Optionally, the reported configuration may also indicate one or more of the following: K, K1 to K P .

[0048] Optionally, K, K1 to K P One or more of them are predefined.

[0049] In another possible design, the first inference information includes: a first inference result and P second inference results; wherein, the p-th second inference result among the P second inference results includes K from the p-th second beam set in the P second beam set. p The information of each beam, wherein the p-th second beam set in the P second beam sets corresponds to the reference signal resource set configured in the p-th second resource configuration in the P second resource configurations; the first inference result includes the information of K' beams in the first beam set, wherein the K' beams are the K' beams in the first beam set excluding the K” beams, and the K” beams are the complete set of the P second beam sets, wherein the first beam set corresponds to the reference signal resource set configured in the first resource configuration; K', K”, K p Both p and p are positive integers, with p taking values ​​from 1 to p.

[0050] Optionally, the reported configuration may also indicate one or more of the following: K', K1 to K P .

[0051] Optionally, K', K1 to K P One or more of them are predefined.

[0052] In another possible design, the first inference information includes: a first inference result and P second inference results; wherein, the first inference information includes K from the first beam set. sum Information about each beam; the first inference result includes the information of the beams in the first beam set excluding the K” beams (K). sum -K”) beam information, the first beam set corresponds to the reference signal resource set configured in the first resource configuration; the K” beams are the K beams in the p-th second beam set obtained by traversing p from 1 to P respectively. p The complete set of P beams, the p-th second inference result among the P second inference results includes K in the p-th second beam set of the P second beam sets. p Information about each beam, the p-th second beam set in the P second beam sets corresponds to the reference signal resource set configured in the p-th second resource configuration in the P second resource configurations, K p Greater than or equal to K p,min K p,min Less than K sum ,K”、K sum K pK p,min Both p and p are positive integers, with p taking values ​​from 1 to p.

[0053] Optionally, the reported configuration may also indicate one or more of the following: K sum and K 1,min To K P,min .

[0054] Optionally, K sum and K 1,min To K P,min One or more of them are predefined.

[0055] In conjunction with the second aspect, in some possible implementations of the second aspect, the reporting configuration further indicates a second reporting period, which is different from the first reporting period, and the first reporting period is an integer multiple of the second reporting period; the method further includes: receiving second inference information at the second reporting time, the second reporting period corresponding to the second reporting period, the second inference information including inference results of reference signal resource sets configured for one or more of the P second resource configurations respectively.

[0056] In conjunction with the second aspect, in some possible implementations of the second aspect, before sending the reported configuration, the method further includes: receiving capability information, which indicates support for determining inference results for a first beam set and multiple fourth beam sets, such as the model supporting the determination of inference results for a first beam set and multiple fourth beam sets, the first beam set corresponding to a reference signal resource set configured by a first resource configuration, each of the multiple fourth beam sets being a subset of the first beam set, and each of the P second resource configurations having a reference signal resource set configured corresponding to one of the multiple fourth beam sets.

[0057] In conjunction with the second aspect, in some possible implementations of the second aspect, before sending the reporting configuration, the method further includes: sending first information, which is used to indicate a first reference signal resource set, the first reference signal resource set corresponding to a first beam set; sending Q second information, which is used to indicate Q fourth reference signal resource sets, each of the Q second information indicating one of the Q fourth reference signal resource sets, the Q fourth reference signal resource sets corresponding one-to-one with the Q fourth beam sets, where Q is a positive integer; wherein each of the Q fourth beam sets is a subset of the first beam set, the reference signal resource set configured by one of the P second resource configurations corresponds to one of the Q fourth beam sets, the first beam set corresponds to the reference signal resource set configured by the first resource configuration, and the first reference signal resource set and the Q fourth reference signal resource sets are used for training the first task.

[0058] In conjunction with the first or second aspect, this reporting configuration is used to configure CSI reporting (or, in other words, to configure CSI reporting). The amount of processing resources occupied by this CSI report is related to the number of reference signal resources in the reference signal resource set configured by the first resource configuration. In other words, the amount of processing resources occupied by this CSI report is independent of the number of reference signal resources in the reference signal resource set configured by any one of the P second resource configurations.

[0059] Therefore, it can avoid the waste of resources caused by duplicate reporting and repeated use of processing resources.

[0060] In this application, processing resources may refer to computing resources, storage resources, etc., and processing resources may be replaced by processing units. The processing resources occupied by the CSI report may be, for example, a CSI processing unit (CPU).

[0061] In conjunction with the first or second aspect, the start time of the CSI report's resource occupancy period is the time when the last reference signal was received before the CSI reference resource, and the end time is the time when the CSI report was submitted.

[0062] In this way, network devices can determine the time that terminal devices occupy CSI processing resources, thereby understanding whether the terminal devices have sufficient processing resources, and thus reasonably configuring CSI reporting for terminal devices.

[0063] Thirdly, a communication method is provided. This method can be applied to a third device, which may be a terminal device or a component deployed within the terminal device, such as circuits or chips inside the terminal device (e.g., a modem chip, or a SoC chip or SIP chip containing a modem core, or a chip for AI or a SoC or SIP chip containing a chip for AI, etc.); or it can be applied to logic modules or software capable of implementing all or part of the functions of the terminal device, etc. For ease of understanding and explanation, the following description uses a terminal device as an example of a third device to illustrate the method.

[0064] The third device in the third aspect may be the same device as the first device in the first aspect, or it may be a different device.

[0065] For example, the method includes: receiving first information, the first information being used to indicate a first beam set; receiving Q pieces of second information, the Q pieces of second information being used to indicate Q fourth beam sets, each of the Q pieces of second information being used to indicate one of the Q fourth beam sets, Q being a positive integer; wherein each of the Q fourth beam sets is a subset of the first beam set, and the first beam set and the Q fourth beam sets are used for training a first task.

[0066] Based on the above scheme, by adding training auxiliary information for subsets during the training data collection process, the training of the first task considers not only the first beam set but also its subsets (i.e., the aforementioned Q fourth beam sets), thereby improving the model's prediction performance for subsets. Furthermore, by utilizing the relationship between the first beam set and its subsets, joint training can be performed for multiple beam sets requiring prediction. For example, only one model can be trained for the first beam set and its subsets, or only one model can be trained for some closely related subsets. Alternatively, a model can be trained first for the first beam set, and then retrained or fine-tuned for its subsets, eliminating the need to train the model from scratch and thus reducing training overhead.

[0067] In conjunction with the third aspect, in some possible implementations of the third aspect, the first information is used to configure the first reference signal resource set, which includes X reference signal resources, and the first beam set includes X beams, with each of the X reference signal resources corresponding to one of the X beams, where X is a positive integer.

[0068] In conjunction with the third aspect, in some possible implementations of the third aspect, Q pieces of second information are used to configure Q sets of fourth reference signal resources, and the Q sets of fourth reference signal resources correspond one-to-one with the Q sets of fourth beams.

[0069] By reusing the configuration signaling of the existing reference signal resource set to indicate the first beam set and Q fourth beam sets, the signaling overhead caused by additional beam configuration can be avoided.

[0070] Optionally, each of the Q fourth reference signal resource sets is a subset of the first reference signal resource set.

[0071] Optionally, the q-th fourth reference signal resource set in the Q fourth reference signal resource sets includes Y. q A reference signal resource, the Y q Y in X reference signal resources and X reference signal resources q Each reference signal resource has a one-to-one mapping relationship, and two reference signal resources with this mapping relationship have the same beam.

[0072] That is, the Q sets of fourth reference signal resources can be subsets of the first reference signal resource set, or they can be non-components of the first reference signal resource set, which is very flexible.

[0073] The subset relationship between the fourth beam set and the first beam set can be indirectly indicated by indicating the relationship between the Q fourth reference signal resource sets and the first reference signal resource set, such as an indication of subset relationship or an indication of mapping relationship.

[0074] In conjunction with the third aspect, in some possible implementations of the third aspect, each of the Q second pieces of information includes a bit map, which includes X indicator bits, each of which corresponds one-to-one with X beams. Each of the X indicator bits included in the qth second piece of information is used to indicate whether the corresponding beam belongs to the fourth beam set indicated by the qth second piece of information.

[0075] The above method can be used to directly indicate the subset relationship between the beams in the fourth beam set and the first beam set.

[0076] In conjunction with the third aspect, in some possible implementations of the third aspect, the method further includes: receiving third information, which indicates that the beam set corresponding to the output result of the first task includes a first beam set and Q fourth beam sets.

[0077] Instructions from the third information allow the terminal device to determine that training for the first task is required on the first beam set and Q fourth beam sets. This enables joint training on multiple beam sets requiring prediction, ensuring that a good model exhibits good predictive performance on subsets.

[0078] In conjunction with the third aspect, in some possible implementations of the third aspect, the method further includes: performing training of the first task based on the first information, the second information, and the third measurement result, wherein the third measurement result is obtained based on measurements of a third set of reference signal resources and includes input data corresponding to the training of the first task.

[0079] Optionally, training for the first task is performed based on the first information, the second information, and the third measurement result, including: measuring a first set of reference signal resources to obtain a first measurement result; measuring Q sets of fourth reference signal resources to obtain Q second measurement results; the first measurement result and the Q second measurement results contain label data corresponding to the training of the first task; obtaining inference results for beam sets indicated by the first information and the second information respectively based on the third measurement result; and performing training for the first task based on the label data and the inference results. Alternatively, it includes: measuring the first set of reference signal resources to obtain a first measurement result; the first measurement result contains label data corresponding to the training of the first task; obtaining inference results for beam sets indicated by the first information and the second information respectively based on the third measurement result; and performing training for the first task based on the label data and the inference results.

[0080] Using the first measurement result and Q second measurement results as label data, joint training can be performed using multiple sets of beams that need to be predicted.

[0081] In conjunction with the third aspect, in some possible implementations of the third aspect, the method further includes: sending capability information, the capability information being used to indicate that the first task supports determining inference results for the first beam set and the Q fourth beam sets.

[0082] By sending capability information, network devices can understand the inference capabilities of the terminal side, and thus configure CSI reports for the terminal devices appropriately.

[0083] In conjunction with the third aspect, in some possible implementations of the third aspect, the method further includes: receiving fourth information for indicating one or more beam set groups, wherein any one of the one or more beam set groups includes a first beam set and at least one of Q fourth beam sets, or includes at least two of Q fourth beam sets.

[0084] By combining the first beam set with one or more fourth beam sets, the terminal can obtain different combinations of the first beam set and Q fourth beam sets, thereby enabling targeted training for different combinations, improving training effectiveness, and saving training costs.

[0085] In conjunction with the third aspect, in some possible implementations of the third aspect, the method further includes: receiving fifth information, which is also used to indicate the performance requirements of the output of the first task.

[0086] Optionally, the performance requirements for the output results of the first beam set can be the same as or different from the performance requirements for the output results of the Q fourth beam sets.

[0087] Therefore, training for the first task can be performed based on this performance requirement, which helps to train a model that meets the performance requirements.

[0088] In conjunction with the third aspect, in some possible implementations of the third aspect, the method further includes: receiving sixth information; the sixth information includes Q association identifiers, which correspond one-to-one with Q fourth beam sets.

[0089] By indicating the association identifier, the terminal device can determine which beam sets have the same beam characteristics, and then use data with the same beam characteristics as the same dataset for training, thereby improving the training effect.

[0090] In conjunction with the third aspect, in some possible implementations of the third aspect, the method further includes: receiving sixth information, which includes an association identifier.

[0091] This sixth piece of information includes an association identifier, which means that the Q fourth beam sets have the same beam characteristics as the first beam set. This reduces the overhead of indicating the association identifier.

[0092] Fourthly, a communication method is provided, which can be applied to a fourth device, which may be a network device or a component deployed in a network device, such as circuits or chips inside the network device (e.g., modem chips, or SoC chips or SIP chips containing modem cores, or chips for AI or SoC or SIP chips containing chips for AI, etc.); or, it can be applied to logic modules or software that can realize all or part of the functions of the network device, etc.

[0093] The fourth device in the fourth aspect and the second device in the second aspect may be the same device or different devices.

[0094] For example, the method includes: sending first information, the first information indicating a first beam set; sending Q second information, the Q second information indicating Q fourth beam sets, each of the Q second information indicating one of the Q fourth beam sets, where Q is a positive integer; wherein each of the Q fourth beam sets is a subset of the first beam set, and the first beam set and the Q fourth beam sets are used for training a first task.

[0095] In conjunction with the fourth aspect, in some possible implementations of the fourth aspect, the first information is used to configure the first reference signal resource set, which includes X reference signal resources, and the first beam set includes X beams, with each of the X reference signal resources corresponding to one of the X beams, where X is a positive integer.

[0096] In conjunction with the fourth aspect, in some possible implementations of the fourth aspect, Q pieces of second information are used to configure Q sets of fourth reference signal resources, and the Q sets of fourth reference signal resources correspond one-to-one with the Q sets of fourth beams. Optionally, each of the Q sets of fourth reference signal resources is a subset of the first reference signal resource set.

[0097] Optionally, the q-th fourth reference signal resource set in the Q fourth reference signal resource sets includes Y. q A reference signal resource, the Y q Y in X reference signal resources and X reference signal resources q Each reference signal resource has a one-to-one mapping relationship, and two reference signal resources with this mapping relationship have the same beam.

[0098] In conjunction with the fourth aspect, in some possible implementations of the fourth aspect, each of the Q second pieces of information includes a bit map, which includes X indicator bits that correspond one-to-one with X beams. Each of the X indicator bits included in the qth second piece of information is used to indicate whether the corresponding beam belongs to the fourth beam set indicated by the qth second piece of information.

[0099] In conjunction with the fourth aspect, in some possible implementations of the fourth aspect, the method further includes: sending third information, which indicates that the beam set corresponding to the output result of the first task includes the first beam set and the Q fourth beam sets.

[0100] In conjunction with the fourth aspect, in some possible implementations of the fourth aspect, the method further includes: receiving capability information, the capability information being used to indicate that the first task supports determining inference results for the first beam set and the Q fourth beam sets.

[0101] In conjunction with the fourth aspect, in some possible implementations of the fourth aspect, the method further includes: sending fourth information for indicating one or more beam set groups, any one of the one or more beam set groups including a first beam set and at least one of Q fourth beam sets, or including at least two of Q fourth beam sets.

[0102] In conjunction with the fourth aspect, in some possible implementations of the fourth aspect, the method further includes: sending fifth information, which is also used to indicate the performance requirements of the output of the first task.

[0103] In conjunction with the fourth aspect, in some possible implementations of the fourth aspect, the method further includes: sending a sixth message; the sixth message includes Q association identifiers, which correspond one-to-one with Q sets of fourth beams.

[0104] In conjunction with the fourth aspect, in some possible implementations of the fourth aspect, the method further includes: sending a sixth message, which includes an association identifier.

[0105] In conjunction with the third or fourth aspect, in some possible implementations, the Q sets of fourth beams share the same association identifier as the first beam set.

[0106] In conjunction with the third or fourth aspect, in some possible implementations, the association identifiers of the Q fourth beam sets are different from those of the first beam set.

[0107] Fifthly, an apparatus is provided. This apparatus may include functional modules corresponding to each of the methods / operations / steps / actions described in any one of the first to fourth aspects, and in any possible implementation of any one aspect. The module may be a hardware circuit, software, or a combination of hardware circuitry and software implementation.

[0108] In one design, the device may include a processing module and a communication module. The communication module is used to perform the sending and receiving actions performed by the first device as described in the first aspect and any possible implementation thereof, while the processing module is used to perform the processing-related actions performed by the first device in the methods described in the first aspect and any possible implementation thereof.

[0109] In one design, the device may include a processing module and a communication module. The communication module is used to perform the sending and receiving actions performed by the third device as described in the third aspect and any possible implementation thereof, while the processing module is used to perform the processing-related actions performed by the third device in the methods described in the third aspect and any possible implementation thereof.

[0110] In one design, the device can be a terminal device, or a device, module, circuit or chip configured in the terminal device, or a device that can be used in conjunction with the terminal device, such as a host or cloud server of an over-the-top (OTT) system.

[0111] In one design, the device may include a processing module and a communication module. The communication module is used to perform the sending and receiving actions performed by the second device in the methods described in the second aspect and any possible implementation thereof, while the processing module is used to perform processing-related actions performed by the second device in the methods described in the second aspect and any possible implementation thereof.

[0112] In one design, the device may include a processing module and a communication module. The communication module is used to perform the sending and receiving actions performed by the fourth device in the methods described in the fourth aspect and any possible implementation thereof, while the processing module is used to perform processing-related actions performed by the fourth device in the methods described in the fourth aspect and any possible implementation thereof.

[0113] In one design, the device can be a network device, or a device, module, circuit, or chip configured in the network device, or a device that can be used in conjunction with the network device, such as an intelligent network element with a radio access network (RAN) intelligent controller (RIC) deployed thereon.

[0114] A sixth aspect provides an apparatus including a processor and a storage medium storing instructions that, when executed by the processor, cause a method as described in any of the first to fourth aspects, and in any possible implementation thereof, to be implemented.

[0115] A seventh aspect provides an apparatus including a processing circuit for processing data and / or information such that a method as described in any of the first to fourth aspects, and in any possible implementation thereof, is implemented.

[0116] The processing circuit may include one or more processors, or all or part of the circuitry in one or more processors used for control or processing functions.

[0117] Optionally, the apparatus may further include a memory for storing programs or instructions, and the processor for running the programs or instructions to implement methods such as any of the first to fourth aspects, and any possible implementation of any of the aspects.

[0118] Optionally, the device may also include the transceiver circuit, or an input / output interface.

[0119] Eighthly, a chip is provided, including processing circuitry for running programs or instructions to implement methods such as any one of the first to fourth aspects, and any possible implementation thereof.

[0120] Optionally, the chip may further include a memory for storing programs or instructions.

[0121] Optionally, the chip may also include transceiver circuitry, or input / output interfaces.

[0122] A ninth aspect provides a computer-readable storage medium comprising instructions that, when executed by a processor, cause a method as described in any of the first to fourth aspects, and in any possible implementation thereof, to be implemented.

[0123] In a tenth aspect, a computer program product is provided, the computer program product comprising computer program code or instructions, which, when executed, cause a method as described in any of the first to fourth aspects, and in any possible implementation thereof, to be implemented.

[0124] Eleventhly, a communication system is provided, which includes a first device and a second device.

[0125] Optionally, the first device is used to perform the method in the first aspect and any possible implementation of the first aspect, and the second device is used to perform the method in the second aspect and any possible implementation of the second aspect.

[0126] In a twelfth aspect, a communication system is provided, the communication system comprising a third device and a fourth device.

[0127] Optionally, the third device is used to perform the method in the third aspect and any possible implementation of the third aspect, and the fourth device is used to perform the method in the fourth aspect and any possible implementation of the fourth aspect.

[0128] It should be understood that the fifth to twelfth aspects of this application correspond to the technical solutions of the first to fourth aspects of this application, and the beneficial effects obtained by each aspect and the corresponding feasible implementation are similar, and will not be repeated here. Attached Figure Description

[0129] Figure 1 is a schematic diagram of a communication system applicable to the communication method of this application embodiment;

[0130] Figure 2 is a schematic diagram of another communication system applicable to the communication method of this application embodiment;

[0131] Figure 3 is a schematic diagram of a possible application framework in a communication system;

[0132] Figure 4 is a schematic diagram of another possible application framework in a communication system;

[0133] Figure 5 is a schematic diagram of wide beam and narrow beam;

[0134] Figure 6 is a schematic diagram of a spatial beam prediction use case;

[0135] Figure 7 is a schematic diagram of a time-domain beam prediction use case;

[0136] Figure 8 is a schematic flowchart of the communication method provided in an embodiment of this application;

[0137] Figure 9 is an example of a reporting configuration provided in an embodiment of this application;

[0138] Figure 10 is a schematic diagram illustrating the transmission of first inference information at a first reporting time and the transmission of second inference information at a second reporting time in an embodiment of this application.

[0139] Figure 11 is a schematic flowchart of a communication method provided in another embodiment of this application;

[0140] Figure 12 is a schematic diagram illustrating the time period of processing resource occupation when the first reporting cycle and the second reporting cycle are the same, as shown in an embodiment of this application.

[0141] Figure 13 is a schematic diagram illustrating the time period of processing resource occupation when the first reporting cycle and the second reporting cycle are different, as shown in an embodiment of this application.

[0142] Figure 14 is a schematic flowchart of a communication method provided in another embodiment of this application;

[0143] Figures 15 and 16 are schematic block diagrams of a communication device provided in an embodiment of this application;

[0144] Figure 17 is a schematic diagram of the structure of the AI ​​processor provided in an embodiment of this application. Detailed Implementation

[0145] The technical solutions in this application will now be described with reference to the accompanying drawings.

[0146] To facilitate understanding of the embodiments of this application, the following points will be explained first:

[0147] First, in this application, the indication includes explicit indication (also known as direct indication) and implicit indication (also known as indirect indication). Explicit indication information A means including information A; implicit indication information A means indicating information A through the correspondence between information A and information B, and direct indication information B. The correspondence between information A and information B can be predefined, pre-stored, pre-burned, or pre-configured; or it can refer to indicating information A through information B and preset rules.

[0148] Second, in this application, information C is used to determine information D, which includes both determining information D based solely on information C and determining it based on information C and other information. Furthermore, information C can also be used to determine information D indirectly, for example, in the case where information D is determined based on information E, and information E is determined based on information C.

[0149] Third, in this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates an "or" relationship between the preceding and following related objects, but it does not exclude the possibility of indicating an "and" relationship; the specific meaning can be understood in context. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c; a and b; a and c; b and c; or a and b and c. Here, a, b, and c can be single or multiple.

[0150] Fourth, in this application, the use of prefixes such as "first" and "second" is merely for the purpose of distinguishing and describing different things belonging to the same category, and does not constrain the order, size, or quantity of things. For example, "first information" and "second information" are simply different pieces of information, and there is no temporal sequence, size, or priority relationship between them.

[0151] Fifth, in this application, "send" and "receive" indicate the direction of signal transmission. For example, "send information to the terminal" can be understood as the destination of the information being the terminal, which may include direct transmission via the air interface or indirect transmission via the air interface by other units or modules. "Receive information from a network device" can be understood as the source of the information being the network device, which may include direct reception from the network device via the air interface or indirect reception from the network device via the air interface by other units or modules. "Send" can also be understood as the "output" of the chip interface, and "receive" can also be understood as the "input" of the chip interface.

[0152] In other words, sending and receiving can occur between devices, such as between a terminal and a network device; or they can occur within a device, such as between components, modules, chips, software modules, or hardware modules within a device via a bus, wiring, or interface.

[0153] Sixth, in the embodiments of this application, "when," "if," and "if" all refer to the device making corresponding processing under certain objective circumstances, and are not limited to a time, nor do they require the device to make a judgment action when it is implemented, nor do they mean that there are other limitations.

[0154] Seventh, in this application, the words "example," "exemplarily," "for example," or "such as" are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "example," "exemplarily," "for example," or "such as" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the words "example," "exemplarily," "for example," or "such as" is intended to present the relevant concepts in a specific manner.

[0155] Eighth, the numbers involved in this application, such as p, can start from 1, or start from 0, or start from any other value. This application does not limit the range of values ​​for each number.

[0156] Ninth, in this application, the terms "of," "corresponding (relevant)," and "corresponding" can sometimes be used interchangeably. It should be noted that when their distinction is not emphasized, their intended meanings are consistent. Furthermore, "corresponding to" in this application can also be replaced with "as" or "there is a corresponding relationship between xx and xx," etc. Similarly, "including" in this application can also be replaced with "as" or "is."

[0157] Tenth, in this application, time can be interchanged with moment, and moment can be a second (s), millisecond (ms), microsecond (us), frame, subframe, slot, symbol, or at least one continuous symbol, etc. This application does not impose any restrictions on this.

[0158] For example, the time when the last reference signal was received could refer to the moment when the first symbol or the last symbol of the last reference signal was received. The reporting time of the CSI report could refer to the start or end time of the CSI report.

[0159] Eleventh, in this application, the first task can be a CSI reporting task. For example, the first task can be a CSI reporting task for inference. A CSI reporting task for inference can be understood as a CSI reporting task that uses an AI model for inference. For example, the first task is to report the CSI report corresponding to the beam prediction results. The first task is a task related to the AI ​​model; "task" can be interchanged with "model" or "function".

[0160] The technical solutions provided in this application can be applied to various communication systems, such as 5th generation (5G) or new radio (NR) systems, frequency division duplex (FDD) systems, time division duplex (TDD) systems, wireless local area network (WLAN) systems, satellite communication systems, future communication systems, or integrated systems of multiple systems. The technical solutions provided in this application can also be applied to device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-to-machine (M2M) communication, machine-type communication (MTC), and Internet of Things (IoT) communication systems or other communication systems.

[0161] In a communication system, one network element can send signals to or receive signals from another network element. These signals can include information, signaling, or data. The term "network element" can also be replaced by an entity, network entity, device, communication equipment, communication module, node, communication node, etc. This disclosure uses a network element as an example. For instance, a communication system can include at least one terminal device and at least one network device. The network device can send downlink signals to the terminal device, and / or the terminal device can send uplink signals to the network device. It is understood that the terminal device in this disclosure can be replaced by a first network element, and the network device can be replaced by a second network element, both performing the corresponding methods described in this disclosure.

[0162] In wireless communication networks, such as mobile communication networks, the services supported by the networks are becoming increasingly diverse, thus requiring increasingly diverse demands. For example, networks need to support ultra-high speeds, ultra-low latency, and / or massive connectivity. This characteristic makes network planning, network configuration, and / or resource scheduling increasingly complex. Furthermore, as network functions become more powerful, such as supporting higher spectrum, supporting higher-order multiple-input multiple-output (MIMO) technologies, supporting beamforming, and / or supporting beam management, network energy saving has become a hot research topic. These new demands, new scenarios, and new characteristics bring unprecedented challenges to network planning, operation, and efficient operation. To meet these challenges, artificial intelligence (AI) technology can be introduced into wireless communication networks to achieve network intelligence. To support AI technology in wireless networks, AI nodes may also be introduced. AI nodes can be the AI ​​network elements shown in Figure 2 below or the AI ​​modules shown in Figure 3.

[0163] Figure 1 is a schematic diagram of a communication system applicable to the communication method of this application embodiment. As shown in Figure 1, the communication system 100A may include at least one access network device, such as access network device 110 shown in Figure 1; the communication system 100A may also include at least one terminal device, such as terminal device 120 and terminal device 130 shown in Figure 1. Access network device 110 and terminal devices (such as terminal device 120 and terminal device 130) can communicate via a wireless link. The communication devices in this communication system, for example, access network device 110 and terminal device 120, can communicate via multi-antenna technology.

[0164] Figure 2 is a schematic diagram of another communication system applicable to the communication method of this application embodiment. Compared with the communication system 100A shown in Figure 1, the communication system 100B shown in Figure 2 further includes an AI network element 140. The AI ​​network element 140 is used to perform AI-related operations, such as building datasets or AI models. The AI ​​network element can also be simply referred to as an intelligent network element. In this disclosure, the AI ​​model can be simply referred to as a model.

[0165] In one possible implementation, access network device 110 can send data related to the training of the AI ​​model to AI network element 140, whereby AI network element 140 constructs a dataset and trains the AI ​​model. For example, the data related to the training of the AI ​​model may include data reported by terminal devices. AI network element 140 can send the results of operations related to the AI ​​model to access network device 110, and then forward them to terminal devices via access network device 110. For example, the results of operations related to the AI ​​model may include at least one of the following: a trained AI model, model evaluation results, or test results, etc. Exemplarily, a portion of the trained AI model may be deployed on access network device 110, and another portion on terminal devices 120 and / or 130. Alternatively, the trained AI model may be deployed on access network device 110. Or, the trained AI model may be deployed on terminal devices 120 and / or 130.

[0166] It should be understood that Figure 2 is only used as an example of the AI ​​network element 140 being directly connected to the access network device 110. In other scenarios, the AI ​​network element 140 can also be connected to a terminal device (such as terminal device 120 and / or terminal device 130); or, the AI ​​network element 140 can be connected to both the access network device 110 and the terminal device (such as terminal device 120 and / or terminal device 130) simultaneously; or, the AI ​​network element 140 can also be connected to the access network device 110 through a third-party network element; the AI ​​network element 140 can also be set as a module in the access network device 110 or the terminal device (such as terminal device 120 and / or terminal device 130) shown in Figure 1. This application embodiment does not limit the connection relationship between the AI ​​network element and other network elements.

[0167] It should be noted that Figures 1 and 2 are simplified schematic diagrams for ease of understanding. For example, the communication system may also include other devices, such as wireless relay devices and / or wireless backhaul devices, which are not shown in Figures 1 and 2. In practical applications, the communication system may include multiple access network devices and multiple terminal devices. The embodiments of this application do not limit the number of access network devices and terminal devices included in the communication system.

[0168] In the embodiments of this application, the terminal device may also be referred to as user equipment (UE), access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent, or user apparatus.

[0169] Terminal devices can be devices that provide voice / data, such as handheld devices with wireless connectivity, in-vehicle devices, etc. Currently, examples of terminals include: mobile phones, tablets, laptops, PDAs, mobile internet devices (MIDs), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminals in industrial control, wireless terminals in self-driving, wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, cellular phones, cordless phones, session initiation protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), handheld devices with wireless communication capabilities, computing devices or other processing devices connected to a wireless modem, wearable devices, terminal devices in 5G networks, or future public land mobile communication networks. Terminal devices in a network (PLMN), etc., are not limited to this in the embodiments of this application.

[0170] By way of example and not limitation, in this embodiment, the terminal device can also be a wearable device. Wearable devices, also known as wearable smart devices, are a general term for devices that utilize wearable technology to intelligently design and develop everyday wearables, such as glasses, gloves, watches, clothing, and shoes. Wearable devices are portable devices that are worn directly on the body or integrated into the user's clothing or accessories. Wearable devices are not merely hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are feature-rich, large in size, and can achieve complete or partial functions without relying on a smartphone, such as smartwatches or smart glasses, as well as those that focus on a specific type of application function and require the use of other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0171] In this embodiment, the device for implementing the functions of the terminal device can be the terminal device itself, or it can be any device capable of supporting the terminal device in implementing those functions, such as a chip system. This device can be installed in or used in conjunction with the terminal device. In this embodiment, the chip system can be composed of chips or may include chips and other discrete components. This embodiment only uses the terminal device as an example to illustrate the device for implementing the functions of the terminal device, and does not constitute a limitation on the solution of this embodiment.

[0172] The network device in this application embodiment can be a device for communicating with a terminal device. This network device can include an access network device (i.e., an access network node) or a radio access network device, such as a base station. In this application embodiment, the radio access network device can refer to a radio access network (RAN) node (or device) that connects the terminal device to the wireless network. A base station can broadly encompass, or be replaced by, various names including: NodeB, evolved NodeB (eNB), next-generation NodeB (gNB), relay station, access point, transmitting and receiving point (TRP), transmitting point (TP), master station, auxiliary station, motor slide retainer (MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), radio unit (RU), positioning node, etc. A base station can be a macro base station, micro base station, relay node, donor node, or similar entities, or combinations thereof. A base station can also refer to a communication module, modem, or chip installed within the aforementioned equipment or apparatus. A base station can also be a mobile switching center, equipment performing base station functions in D2D, V2X, and M2M communications, network-side equipment in future communication networks, or equipment performing base station functions in future communication networks. A base station can support networks using the same or different access technologies. Optionally, a RAN node can also be a server, wearable device, vehicle, or in-vehicle equipment. For example, the access network equipment in vehicle-to-everything (V2X) technology can be a roadside unit (RSU). The embodiments of this application do not limit the specific technologies or equipment forms used in the network equipment.

[0173] Base stations can be fixed or mobile. For example, a helicopter or drone can be configured to act as a mobile base station, and one or more cells can move depending on the location of the mobile base station. In other examples, a helicopter or drone can be configured as a device to communicate with another base station.

[0174] In some deployments, the access network equipment mentioned in the embodiments of this application may be a device including a CU, or a DU, or a device including both CU and DU, or a device with a control plane CU node (central unit-control plane (CU-CP)) and a user plane CU node (central unit-user plane (CU-UP)) and a DU node. For example, the access network equipment may include gNB-CU-CP, gNB-CU-UP, and gNB-DU.

[0175] In some deployments, multiple RAN nodes collaborate to assist terminals in achieving wireless access, with different RAN nodes each implementing some of the base station's functions. For example, RAN nodes can be CUs, DUs, CU-CPs, CU-UPs, or RUs. CUs and DUs can be configured separately or included in the same network element, such as a BBU. RUs can be included in radio frequency equipment or radio frequency units, such as RRUs, AAUs, or RRHs.

[0176] RAN nodes can support one or more types of fronthaul interfaces, each corresponding to a DU and RU with different functions. If the fronthaul interface between the DU and RU is a Common Public Radio Interface (CPRI), the DU is configured to implement one or more baseband functions, and the RU is configured to implement one or more radio frequency functions. If the fronthaul interface between the DU and RU is another type of interface, relative to CPRI, it moves some downlink and / or uplink baseband functions—for example, for downlink, precoding, digital beamforming, or one or more of inverse fast Fourier transform (IFFT) / adding a cyclic prefix (CP)—from the DU to the RU; and for uplink, digital beamforming, or one or more of fast Fourier transform (FFT) / removing CP—from the DU to the RU. In one possible implementation, this interface can be an enhanced common public radio interface (eCPRI). Under the eCPRI architecture, the splitting methods between DU and RU are different, corresponding to different types (category, Cat) of eCPRI, such as eCPRI Cat A, B, C, D, E, F.

[0177] Taking eCPRI Cat A as an example, for downlink transmission, layer mapping is used as the dividing line. The DU is configured to implement one or more functions preceding layer mapping (i.e., coding, rate matching, scrambling, modulation, and layer mapping itself), while other functions following layer mapping (e.g., resource element (RE) mapping, digital beamforming, or one or more of IFFT / CP addition) are implemented in the RU. For uplink transmission, de-RE mapping is used as the dividing line. The DU is configured to implement one or more functions preceding de-mapping (i.e., decoding, rate matching de-matching, descrambling, demodulation, inverse discrete Fourier transform (IDFT), channel equalization, and de-RE mapping itself), while other functions following de-mapping (e.g., digital BF or FFT / CP removal) are implemented in the RU. It is understood that descriptions of the functions of the DU and RU corresponding to various types of eCPRI can be found in the eCPRI protocol and will not be elaborated upon here.

[0178] In one possible design, the processing unit in the BBU used to implement baseband functions is called the baseband high (BBH) unit, and the processing unit in the RRU / AAU / RRH used to implement baseband functions is called the baseband low (BBL) unit.

[0179] In different systems, CU (or CU-CP and CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an open RAN (ORAN) architecture, CU can also be called open CU (open-CU, O-CU), DU can also be called open DU (open-DU, O-DU), CU-CP can also be called open CU-CP (open-CU-CP) O-CU-CP, CU-UP can also be called open CU-UP (open-CU-UP, O-CU-UP), and RU can also be called open RU (open-RU, O-RU). Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software modules and hardware modules.

[0180] In this embodiment, the apparatus for implementing the functions of a network device can be a network device itself; it can also be an apparatus capable of supporting the network device in implementing those functions, such as a chip system, hardware circuit, software module, or a hardware circuit plus a software module. This apparatus can be installed in the network device or used in conjunction with the network device. In this embodiment, the example of a network device being used to implement the functions of a network device is provided only and does not constitute a limitation on the solutions described in this embodiment.

[0181] Network devices and / or terminal devices can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; and they can also be deployed in the air on airplanes, balloons, and satellites. This application does not limit the scenario in which the network devices and terminal devices are located. Furthermore, terminal devices and network devices can be hardware devices, or software functions running on dedicated hardware or general-purpose hardware, such as virtualization functions instantiated on a platform (e.g., a cloud platform), or entities that include dedicated or general-purpose hardware devices and software functions. This application does not limit the specific form of the terminal devices and network devices.

[0182] Optionally, the AI ​​node can be deployed in one or more of the following locations within the communication system: access network equipment, terminal equipment, or core network elements. Alternatively, the AI ​​node can also be deployed independently, for example, in a location other than any of the aforementioned devices, such as in the host or cloud server of an over-the-top (OTT) system. The AI ​​node can communicate with other devices in the communication system, which can be, for example, one or more of the following: access network equipment, terminal equipment, or core network elements.

[0183] It is understood that this application does not limit the number of AI nodes. For example, when there are multiple AI nodes, these nodes can be divided based on function, such as different AI nodes being responsible for different functions.

[0184] It can also be understood that AI nodes can be AI network elements or AI modules. AI nodes can be independent devices, or they can be integrated into the same device to implement different functions. Alternatively, they can be network elements in hardware devices, software functions running on dedicated hardware, or virtualization functions instantiated on a platform (e.g., a cloud platform). This application does not limit the specific form of the AI ​​nodes described above.

[0185] Figure 3 illustrates a possible application framework in a communication system. As shown in Figure 3, network elements in the communication system are connected via interfaces (e.g., next-generation (NG) interfaces, Xn interfaces) or air interfaces. The NG interface is the interface between the radio access network and the 5G core network. The Xn interface is the interface between access network devices, and the air interface is the interface between access network devices and terminal devices. These network element nodes, such as core network devices, RAN nodes, terminal devices, or one or more devices in operation administration and maintenance (OAM), are equipped with one or more AI modules (only one is shown in Figure 3 for clarity). The access network node can be a single RAN node or can include multiple RAN nodes, for example, including CU and DU. The CU and / or DU can also be equipped with one or more AI modules. Optionally, the CU can be further divided into CU-CP and CU-UP. One or more AI models are configured in the CU-CP and / or CU-UP. For example, the CU and DU are connected via an F1 interface. The CU and CU are connected via an Xn interface.

[0186] AI modules are used to implement corresponding AI functions. AI modules deployed in different network elements can be the same or different. Depending on the parameter configuration, the AI ​​module can implement different functions. The AI ​​module model can be configured based on one or more of the following parameters: structural parameters (e.g., at least one of the following: number of neural network layers, neural network width, inter-layer connections, neuron weights, neuron activation function, or bias in the activation function), input parameters (e.g., at least one of the following: type of input parameter, input dimension, number of input ports), or output parameters (e.g., at least one of the following: type of output parameter, output dimension, number of output ports). The bias in the activation function can also be called the bias of the neural network. Input dimension can refer to the size of an input data set; for example, when the input data is a sequence, the input dimension corresponding to that sequence can indicate the length of the sequence. The number of input ports can refer to the quantity of input data. Similarly, output dimension can refer to the size of an output data set; for example, when the output data is a sequence, the output dimension corresponding to that sequence can indicate the length of the sequence. The number of output ports can refer to the quantity of output data.

[0187] An AI module can have one or more models. A model can infer an output, which includes one or more parameters. The learning, training, or inference processes of different models can be deployed on different nodes or devices, or they can be deployed on the same node or device.

[0188] Network devices can be network devices equipped with one or more AI modules. These network devices can include one or more devices in the core network, RAN, or OAM as shown in Figure 3. For example, the AI ​​module can be a RAN intelligent controller (RIC) as shown in Figure 4, such as a near-real-time RIC (near-RT RIC) or a non-real-time RIC (non-RT RIC). For instance, a near-real-time RIC is located in a RAN node (e.g., in a CU or DU), while a non-real-time RIC is located in the OAM, a cloud server, a core network device, or other access network devices. The RIC can obtain subsets from multiple terminal devices from RAN nodes (e.g., CU, CU-CP, CU-UP, DU, and / or RU), reassemble them into a dataset, and train based on the dataset. Exemplarily, near-real-time RICs and non-real-time RICs can also be configured as separate network elements, and access network devices can be either near-real-time or non-real-time RICs.

[0189] Figure 4 illustrates another possible application framework in a communication system. In addition to access network nodes (CU, DU, and RU are shown in the figure) and terminals, the communication system shown in Figure 4 also includes an RIC (Regulator-Integrated Circuit). For example, the RIC could be the AI ​​module shown in Figure 3, which can be used to implement AI-related functions. The RIC includes near-real-time RICs and non-real-time RICs. Non-real-time RICs primarily process non-real-time information, such as data that is not sensitive to latency, with latency on the order of seconds. Real-time RICs primarily process near-real-time information, such as data that is relatively sensitive to latency, with latency on the order of tens of milliseconds.

[0190] Near real-time (NRT) RICs are used for model training and inference. For example, they are used to train AI models and then use those models for inference. NRT RICs can obtain network-side and / or terminal-side information from RAN nodes (e.g., CUs, CU-CPs, CU-UPs, DUs, and / or RUs) and / or terminals. This information can be used as training data or inference data. Optionally, the NRT RIC can deliver inference results to RAN nodes and / or terminals. Optionally, inference results can be exchanged between CUs and DUs, and / or between DUs and RUs. For example, the NRT RIC delivers inference results to a DU, which then forwards them to an RU.

[0191] Non-real-time RICs are also used for model training and inference. For example, they can be used to train AI models and then use those models for inference. Non-real-time RICs can obtain network-side and / or terminal-side information from RAN nodes (e.g., one or more of CU, CU-CP, CU-UP, DU, or RU) and / or terminals. This information can be used as training data or inference data, and the inference results can be delivered to the RAN nodes and / or terminals. Optionally, inference results can be exchanged between CUs and DUs, and / or between DUs and RUs; for example, a non-real-time RIC delivers inference results to a DU, which then forwards them to an RU.

[0192] Near real-time RICs and non-real-time RICs can also be configured as separate network elements. Alternatively, near real-time RICs and non-real-time RICs can also be part of other devices. For example, near real-time RICs can be set in RAN nodes (e.g., CU, DU), while non-real-time RICs can be set in OAM, cloud servers, core network devices, or other network devices.

[0193] To facilitate understanding of the embodiments of this application, the following is a brief introduction to several concepts involved in this document.

[0194] 1. CSI Feedback:

[0195] In LTE and NR communication systems, network devices need to acquire downlink CSI to determine the resources, modulation and coding scheme (MCS), precoding, and other configurations for scheduling downlink data channels for terminal devices. In time division duplex (TDD) systems, due to the reciprocity of uplink and downlink channels, network devices can acquire uplink CSI by measuring uplink reference signals and then infer a more accurate downlink CSI, for example, using the uplink CSI as the downlink CSI. In frequency division duplex (FDD) systems, uplink and downlink reciprocity cannot be guaranteed. The downlink CSI is acquired by the terminal device by measuring downlink reference signals, such as the channel state information reference signal (CSI-RS) or the synchronizing signal block (SSB) (i.e., the synchronization signal (SS) / physical broadcasting channel (PBCH) block, or SS / PBCH block). Therefore, the terminal device needs to generate a CSI report according to the protocol predefined method or the base station configuration and feed the CSI back to the base station so that it can acquire the downlink CSI.

[0196] In the NR protocol, the downlink CSI configuration and reporting process is as follows: The network device sends a CSI reporting configuration (CSI-ReportConfig) to the terminal device, specifying the reporting type (reportConfigType), reporting quantity (reportQuantity), etc. The reporting type can be periodic, semi-persistent, or aperiodic; the reporting quantity can be a rank indicator (RI), channel quality indicator (CQI), precoding matrix indicator (PMI), reference signal received power (RSRP), etc.

[0197] For example, the network device sends a CSI-RS to the terminal device. The terminal device performs channel and interference measurements based on the CSI-RS to obtain measurement results. Based on the measurement results, the terminal device determines the various reporting parameters to be configured and reports downlink CSI to the network device. This downlink CSI includes RI, CQI, PMI, RSRP, etc., measured by the terminal device. If the reporting type in CSI-ReportConfig is configured as periodic, the terminal device reports periodically according to the period specified in the radio resource control (RRC) signaling, and the network device does not need to send signaling to trigger the reporting each time. If the reporting type in CSI-ReportConfig is configured as semi-persistent, the initial reporting of downlink CSI by the terminal device needs to be triggered by signaling, and once triggered, it reports periodically according to the specified period. When CSI is reported on the physical uplink control channel (PUCCH), network devices trigger the report using medium / media access control (MAC) control element (CE) signaling. When CSI is reported on the physical uplink shared channel (PUSCH), network devices can trigger the report using downlink control information (DCI). If the reporting type in CSI-ReportConfig is configured as aperiodic, then DCI is required to trigger the report. Semi-persistent CSI reporting triggering is more complex.

[0198] The protocol configures CSI measurement resources (CSI-RS) and CSI reports via RRC signaling, corresponding to CSI resource configuration (CSI-ResourceConfig) and CSI report configuration (CSI-ReportConfig), respectively. CSI-ReportConfig indicates the identifier (CSI-ResourceConfigId) of the resource configuration used for the CSI measurement in that report. CSI-RS configuration is divided into three levels: CSI-ResourceConfig, CSI-RS resource set, and CSI-RS resource. CSI-RS includes two types: zero-power (ZP) CSI-RS and non-zero-power (NZP) CSI-RS. ZP CSI-RS is mainly used for interference estimation, while NZP CSI-RS can be used for CSI measurement and estimation of the transmission channel, interference estimation, and Layer 1 (L1) RSRP (L1-RSRP) measurement. Taking the configuration of NZP-CSI-RS as an example, each CSI-ResourceConfig can configure S (S≥1) non-zero power CSI-RS resource sets (NZP-CSI-RS-ResourceSet) (given by the higher-layer signaling parameter nzp-CSI-RS-ResourceSetList, indicating S NZP-CSI-RS-ResourceSetIds). For the NZP-CSI-RS-ResourceSets s among the S NZP-CSI-RS-ResourceSets, K can be configured. s (K s ≥1) Non-zero power CSI-RS resources (NZP-CSI-RS-Resource) (given by the higher-layer signaling parameter nzp-CSI-RS-Resources, indicating K) s The time-domain behavior of an NZP-CSI-RS-Resource is indicated by the high-level parameter "resourceType" in the CSI-ResourceConfig, which can be periodic, aperiodic, or semi-persistent. All NZP-CSI-RS-Resources associated with the same CSI-ResourceConfig have the same time-domain behavior. The time-frequency resources used to transmit an NZP-CSI-RS-Resource are configured through the high-level signaling parameter "resourceMapping".

[0199] The maximum number of active CSI-RS resources that a terminal device can simultaneously receive is constrained by the terminal device's capabilities. Since a CSI-RS resource can be referenced by multiple CSI reporting configurations, the protocol specifies a CSI-RS counting mechanism to ensure that the terminal device and network device align on whether the number of currently received CSI-RS resources exceeds the terminal device's capabilities. If a CSI-RS resource is referenced N times by one or more CSI reporting configurations, the count is N.

[0200] Furthermore, the NR protocol also specifies the CSI processing guidelines for terminal devices. The number N of available CSI processing units (CPUs) that a terminal device can report is... CPU This means that the terminal device simultaneously supports N CPU The calculation of CSI reports. In a given time domain symbol, if the calculation of a CSI report occupies L CPUs, then the terminal device has (N... CPU -L) unused CPUs. For a given time-domain symbol, there are N... CPU If -L CPUs are not occupied, and N CSI reports require them to occupy their respective CPUs starting from this time domain symbol, then the number of CPUs corresponding to the nth CSI report is... If n = 0, ..., N-1, then the terminal device does not need to update (NM) lowest priority CSI reports, where 0 ≤ M ≤ N, and M is a subset of the CSI reports that satisfy the condition... The maximum value. That is, when there are not enough unused CPUs to process all CSI reports, the terminal device can prioritize and not process some CSI reports. The number of CPUs required to process each CSI report. This is related to the configured reporting volume, the number of reference signal resources used for channel measurements, and the number of prediction windows. For example:

[0201] When the reported quantity is configured as 'none', then

[0202] When the reported quantity is configured as RSRP, then

[0203] When the reported volume is configured as PMI, except in the following cases... K s This refers to the number of CSI-RS resources in the CSI-RS resource set used for channel measurements.

[0204] • When the reporting quantity is configured as PMI, and the codebook type is 'Type II-Doppler-Version 18' or 'Type II-Doppler-PortSelection-Version 18', for a non-periodic resource set configured with K resources, when K=12, When K < 12 This is related to the capabilities of the terminal device. For a configured periodic or semi-persistent resource set, when N4 = 1, When N4>1 Y2 is related to the capabilities of the terminal device, and N4 is the number of predicted instances.

[0205] • When L sub-reporting configurations are configured (or, in other words, sub-reporting configurations, report sub-configurations, reporting sub-configurations, etc.), It is the total number of resources reported by the i-th sub-report.

[0206] In addition, the CPU continuously occupies a certain number of symbols for processing each CSI report. The protocol specifies that when the reporting type (reportConfigType) is not set to 'none', the number of CPU symbols occupied is determined according to the following rules:

[0207] • CPU time consumed by periodic or semi-persistent CSI reports (excluding initial semi-persistent CSI reports on the PUSCH after a PDCCH trigger report, and excluding semi-persistent CSI reports with codebook types 'typeII-Doppler-r18' or 'typeII-Doppler-PortSelection-r18'): from the first symbol of the earliest reference signal resource used for channel measurements until the last symbol of the PUSCH / PUCCH carrying the report. The most recent measurement resource is no later than the corresponding CSI reference resource.

[0208] • CPU time consumed by non-periodic CSI reports: from the first symbol after the PDCCH that triggers the CSI report to the last symbol of the PUSCH that carries the report.

[0209] • CPU time occupied by the initial semi-persistent CSI report on the PUSCH after PDCCH triggering: from the first symbol after PDCCH until the last symbol of the PUSCH carrying the report.

[0210] • CPU time consumed by a semi-persistent CSI report with codebook type 'typeII-Doppler-r18' or 'typeII-Doppler-PortSelection-r18': no ​​later than the Kth digit of the CSI reference resource before the PUSCH carrying the report. p The sequence begins with the first symbol of the latest CSI-RS timing and ends with the last symbol of the PUSCH carrying the report. p It is related to the capabilities of the terminal device.

[0211] 2. Beam Measurement:

[0212] Beam management refers to the process by which terminal and network devices periodically identify the optimal beam. The optimal beam can be the beam that maximizes received or transmitted power. For example, if a receiver uses different receive beams to receive a reference signal (RS), the optimal beam can include the beam with the highest measured value of the corresponding reference signal among multiple different receive beams. Similarly, if a transmitter uses different transmit beams to transmit a signal, the optimal beam can include the beam with the highest measured value of the corresponding reference signal when the transmitted reference signal arrives at the receiver among multiple transmit beams. The measured value of the reference signal can be, for example, the measured reference signal received power (RSRP), signal-to-interference-plus-noise ratio (SINR), or other possible estimates.

[0213] A beam is a communication resource. In the NR protocol, beams can be represented as spatial filters, or spatial parameters. The beam used to transmit signals can be called the transmission beam (Tx beam), or a spatial domain transmit filter or spatial domain transmit parameter; the beam used to receive signals can be called the reception beam (Rx beam), or a spatial domain receiver filter or spatial domain receive parameter. The transmission beam refers to the distribution of signal strength in different directions in space after the signal is transmitted through the antenna, while the reception beam refers to the distribution of signal strength in different directions in space of the wireless signal received from the antenna.

[0214] Beams can be identified by their identifier (ID). For example, a beam ID can be a Channel State Information Reference Signal Resource Indicator (CSI-RS, CRI), an SSB Resource Indicator (SSBRI), or a bit in a bitmap corresponding to the beam, or an index of the beam within a beam set. For instance, the number of bits in the bitmap is equal to the total number of beams associated with the network device in a single beam management inference task. Beams can be categorized as wide beams and narrow beams. A wide beam refers to a beam with a relatively large radiation range of the transmitting or receiving antenna when transmitting or receiving signals. Wide beams are typically used in applications requiring broadcasting signals to a large area or providing wide coverage. Wide beams can provide a wider coverage area, but the signal strength is relatively weaker. A narrow beam refers to a beam with a relatively small radiation range of the transmitting or receiving antenna. Narrow beams are typically used in applications requiring focusing signals onto a specific target or area. Narrow beams can provide higher signal strength and greater directivity, but the coverage area is relatively small.

[0215] The above measurements (i.e., measurement results) can also be called information characterizing the channel state, or CSI measurements.

[0216] In this application, CSI may include one or more of the following: rank indication (RI) information, channel quality indicator (CQI) information, precoding matrix indicator (PMI), or reference signal receiver power (RSRP), such as layer 1 reference signal receiver power (L1-RSRP), reference signal receiver quality (RSRQ), and signal to interference plus noise ratio (SINR).

[0217] In this application, reference signal resources are used to carry reference signals, which refer to signals known to the terminal device. Exemplarily, the reference signal includes one or more of the following: synchronizing signal block (SSB), channel state information reference signal (CSI-RS), tracking reference signal (TRS), phase-tracking reference signal (PTRS), and positioning reference signal (PRS). In other words, the reference signal can be replaced by any one of the known signal, downlink signal, SSB, CSI-RS, TRS, PTRS, and PRS. The reference signal can be used for channel measurement or channel estimation, etc. The reference signal resources can be used to configure the transmission attributes of the reference signal, such as one or more of time-frequency resource locations, port mapping relationships, power factors, and scrambling codes, etc., without specific limitations. The transmitting device can transmit the reference signal based on the reference signal resources, and the receiving device can receive the reference signal based on the reference signal resources. The reference signal resources may include beams, or have a corresponding relationship with beams. Furthermore, the reference signal resource may also include time-domain resources and / or frequency-domain resources corresponding to the beam, such as time-frequency resources. The beam can also be referred to as a spatial resource.

[0218] It should be understood that in this application, the reference signal resource corresponding to a beam can be understood as the reference signal resource transmitted using that beam. The beam corresponding to a reference signal resource can be understood as the beam used to transmit that reference signal resource. There is a one-to-one or many-to-one correspondence between reference signal resources and beams. Different reference signal resources can correspond to the same or different beams. In other words, the same beam can be used to transmit different reference signal resources, and different beams can be used to transmit different reference signal resources. Different beams correspond to different reference signal resources. In other words, different beams can be used to transmit different reference signal resources. The same beam can be understood as an energy distribution with the same shape and direction. The same reference signal resource can be understood as a downlink signal resource with the same resource parameters such as time-frequency resource location, port mapping relationship, power factor, and scrambling code. That is, if any one of the resource parameters such as time-frequency resource location, port mapping relationship, power factor, and scrambling code differs between two reference signal resources, then these two reference signal resources can be considered as different reference signal resources. The fact that two reference signal resources correspond to the same beam can be understood as these two reference signal resources being transmitted through the same port, or as these two reference signal resources having the same TCI state, or as these two reference signal resources having the same QCL information.

[0219] In this application, a reference signal can refer to a reference signal resource or a set of reference signal resources. A network device can transmit a corresponding reference signal on the reference signal resource or set of reference signal resources, and a terminal device can measure the corresponding reference signal on the reference signal resource or set of reference signal resources. The configured reference signal resource can be periodic, and correspondingly, the reference signal can be transmitted periodically. For periodically transmitted reference signals, a reference signal can be understood as all reference signal resources corresponding to one period, or as a resource transmission occasion. That is, a reference signal can be replaced by a transmission occasion or a reference signal resource occasion.

[0220] Alternatively, beam can be replaced with signal, downlink beam, transmit beam, transmit beam, thin beam, narrow beam, wide beam, spatial filter, spatial filter, spatial parameters, spatial transmit filter, port, etc.

[0221] To achieve beam management, one possible approach is to reduce beam scanning overhead through methods such as layered scanning. This involves scanning a wide beam first, followed by scanning a portion of narrow beams within the wide beam, thereby reducing overhead. A schematic diagram of wide and narrow beams is shown in Figure 5. As shown in Figure 5, compared to narrow beams, wide beams have a wider beam angle and can transmit signals over a wider range of directions. Conversely, narrow beams have a narrower beam angle and can transmit signals over a smaller range of directions. When the sum of the beam angles of a set of wide beams is the same as the sum of the beam angles of a set of narrow beams, the number of beams included in the wide beam set is less than the number of beams included in the narrow beam set.

[0222] Beam selection is primarily accomplished through reference signals and corresponding beam measurements. Specifically, the reference signal can include a synchronization signal block (SSB, also known as a synchronization signal / physical broadcast channel block (SS / PBCH block) or a synchronization signal block (SS block)), or one or more of the following: CSI-RS. The SSB can be a cell broadcast signal, comprising the primary synchronization signal (PSS), secondary synchronization signal (SSS), physical broadcast channel (PBCH), and demodulation reference signal (DMRS). The SSB can be transmitted periodically according to the cell configuration, and can be considered a wide-beam signal. Correspondingly, the CSI-RS signal can be a UE-level signal, and can be understood as a narrow-beam signal.

[0223] The beam scanning process combines beam measurement, beam reporting, and beam determination to select the optimal beam pair between network devices and terminal devices. Specifically, it finds the most suitable transmit and receive beams through beam scanning, aligning the transmit and receive beam directions to achieve optimal receive signal gain and improve communication quality. For example, the beam scanning process is divided into three stages: P1, P2, and P3, as detailed below:

[0224] P1 Process: Network devices use SSB (Special Signal-Based Beam) for beam scanning, while terminal devices use wide-beam scanning. Network devices can use beam scanning to transmit SSB beams from different directions in a time-division manner, broadcasting synchronization and system messages. Terminal devices can use beam scanning to receive signals and confirm the received beam. Simultaneously, the terminal devices feed back the SSB measurement results to the network devices, which then confirm the transmitted beam. The transmitted and received beams achieve initial alignment. The primary purpose of P1 is to find an initial beam pair between the network devices and the terminal devices.

[0225] P2 Process: The network device uses CSI-RS for beam scanning, while the terminal device uses a fixed receive beam. The network device then scans the area around the SSB beam determined by random access using a narrower CSI-RS beam. The terminal device feeds back the CSI-RS measurement results to the network device via a measurement report, and the network device confirms the optimal transmit beam. The P2 process refines the network device's transmit beam; after the initial beam pair is established, to obtain higher signal gain, a narrower CSI-RS beam than the SSB beam needs to be selected for beam adjustment.

[0226] P3 Process: The network device uses a fixed transmit beam, while the terminal device uses narrow beam scanning. The network device uses a fixed CSI-RS narrow beam, and the terminal device uses beam scanning for signal reception to confirm a more accurate receive beam. The transmit and receive beams are then finally aligned. The P3 process refines the receive beam on the terminal device side, enhancing signal quality through further adjustments.

[0227] As can be seen, SSB or CSI-RS (CSI-RS for Beam Measurement) is used as the reference signal for beam scanning. Therefore, the beam measurement and reporting process is consistent with the CSI configuration and reporting process. For example, in the P2 process, the network device is configured as a CSI-RS resource indicator (CRI)-RSRP through the reportQuantity field in CSI-ReportConfig, instructing the terminal device to report the CRI and the corresponding RSRP.

[0228] 3. Beam indication:

[0229] In this application, the information used to indicate the beam used for transmission can be called beam indication information. Beam indication information can be one or more of the following: beam number (or index, identifier, ID, etc.), identifier of signal resources (e.g., identifier of reference signal resources, such as index or number, where the signal resources can be one or more of uplink signal resources, downlink signal resources, or sidelink signal resources, where the index or number can be absolute, relative, or logical, and can include one or more of the following: group or set index or number, or index or number of resources within a group or set, or index or number of resources), absolute index of the beam, relative index of the beam, logical index of the beam, index of the antenna port corresponding to the beam, index of the antenna port group corresponding to the beam, index of the signal (e.g., downlink signal, uplink signal, or sidelink signal, etc.) corresponding to the beam, time index of the SSB corresponding to the beam, beam pair link (BPL) information, transmit parameters (Tx parameter) corresponding to the beam, and receive parameters (Rx parameter) corresponding to the beam. The beam indication information includes at least one of the following: beam parameter, beam-corresponding transmit weight, beam-corresponding weight matrix, beam-corresponding weight vector, beam-corresponding receive weight, beam-corresponding transmit weight index, beam-corresponding weight matrix index, beam-corresponding weight vector index, beam-corresponding receive weight index, beam-corresponding receive codebook, beam-corresponding transmit codebook, beam-corresponding receive codebook index, and beam-corresponding transmit codebook index. The absolute index of the beam includes, for example, the index of the beam in beam set A, and the relative index of the beam includes, for example, the index of the beam in a subset M of beam set A. The logical index of the beam includes, for example, the bit corresponding to the beam in the bitmap. Beam indication information can also be represented as a transmission configuration indicator (TCI) or a TCI state. A TCI state includes one or more quasi-co-location (QCL) information, each QCL information including the ID of a reference signal (or synchronization signal block) and a QCL type. For example, a terminal device may need to determine the beam to receive the physical downlink shared channel (PDSCH) based on the TCI status indicated by the network device (typically carried by the physical downlink control channel, PDCCH). In this application, the beam index information, i.e., the beam ID, is a typical example of beam indication information. The beam index can also be replaced with other beam indication information that can indicate a beam.In this application, the identification information of the reference signal resource can be replaced with the identification information of the beam corresponding to the reference signal resource or the index information of the beam (such as one or more of absolute index, relative index, or logical index). The reference signal resource can also be replaced with a beam.

[0230] QCL (Quadrature Communication Characteristics) relationships are used to indicate that multiple resources share one or more identical or similar communication characteristics. For multiple resources with a QCL relationship, identical or similar communication configurations can be used. Antenna ports with a QCL relationship have signals with the same parameters, or the parameters of one antenna port (also called QCL parameters) can be used to determine the parameters of another antenna port with a QCL relationship, or two antenna ports have the same parameters, or the parameter difference between two antenna ports is less than a certain threshold. These parameters may include one or more of the following: delay spread, Doppler spread, Doppler shift, average delay, average gain, and spatial Rx parameters. Spatial Rx parameters may include one or more of the following: angle of arrival (AOA), average AOA, AOA spread, angle of departure (AOD), average departure angle AOD, AOD spread, receive antenna spatial correlation parameters, transmit antenna spatial correlation parameters, transmit beam, receive beam, and resource identifier.

[0231] 4. AI Model:

[0232] An AI model is an algorithm or computer program that enables AI functionality. It represents the mapping relationship between the model's input and output. An AI model can be understood as a function model that maps an input of a certain dimension to an output of a certain dimension; its parameters are obtained through machine learning training. For example, f(x) = ax 2 +b is a quadratic function model, which can be viewed as an AI model. a and b correspond to the parameters of this AI model, and can be obtained through machine learning training. An AI model can also be called a model, an AI function, or a feature. One AI function can correspond to one or more AI models.

[0233] AI models can be neural networks, linear regression models, decision tree models, support vector machines (SVM), Bayesian networks, Q-learning models, or other machine learning (ML) models.

[0234] 5. Neural Network (NN):

[0235] Neural networks are a specific implementation of AI or machine learning. According to the general approximation theorem, neural networks can theoretically approximate any continuous function, thus enabling them to learn arbitrary mappings.

[0236] Neural networks can be composed of neural units, which can refer to units represented by x. s The neural network is a computational unit that takes an intercept of 1 as input. A neural network is a network formed by connecting many of these individual neural units together; that is, the output of one neural unit can be the input of another. The input of each neural unit can be connected to the local receptive field of the previous layer to extract features from the local receptive field, which can be a region composed of several neural units.

[0237] Taking neural networks as an example of AI models, the AI ​​model involved in this application can be a deep neural network (DNN). The idea behind DNNs originates from the neuronal structure of the brain. Each neuron performs a weighted summation of its input values ​​and generates an output through a non-linear function. DNNs generally have a multi-layered structure, with each layer containing multiple neurons. The input layer processes the received values ​​through neurons and then passes them to the hidden layers. Similarly, the hidden layers pass the calculation results to the final output layer, producing the final output of the DNN. DNNs typically have more than one hidden layer, which often directly affects the ability to extract information and fit functions. Increasing the number of hidden layers or widening the width of each layer can improve the function fitting ability of the DNN. The weighted values ​​in each neuron are the parameters of the DNN network model. The model parameters are optimized through the training process, enabling the DNN network to extract data features and express mapping relationships. Depending on how the network is constructed, DNNs can include feedforward neural networks (FNNs), convolutional neural networks (CNNs), and recurrent neural networks (RNNs), etc.

[0238] CNNs are neural networks specifically designed to process data with a grid-like structure. For example, time-series data (discrete sampling along the time axis) and image data (two-dimensional discrete sampling) can both be considered grid-like data. CNNs do not use all the input information at once for computation; instead, they use a fixed-size window to extract a portion of the information for convolution operations, which significantly reduces the computational cost of model parameters. Furthermore, depending on the type of information extracted by the window (such as people and objects in an image representing different types of information), each window can use different convolution kernels, allowing CNNs to better extract features from the input data.

[0239] Recurrent Neural Networks (RNNs) are a type of distributed neural network (DNN) that utilizes feedback time-series information. Their input includes the current input value and their own output value from the previous time step. RNNs are well-suited for acquiring temporally correlated sequence features, and are particularly applicable to applications such as speech recognition and channel coding / decoding.

[0240] The characteristic of FNN networks is that neurons in adjacent layers are completely connected to each other, which makes FNNs typically require a large amount of storage space and result in high computational complexity.

[0241] The FNN, CNN, and RNN mentioned above are all constructed based on neurons. As mentioned earlier, each neuron performs a weighted summation operation on its respective input value and generates an output through a nonlinear function. The weights of the weighted summation operation of neurons in a neural network and the nonlinear function are called the parameters of the neural network. The parameters of all neurons in a neural network constitute the parameters of that neural network.

[0242] 6. Dataset:

[0243] A dataset refers to the data used for model training, validation, and testing in machine learning. The quantity and quality of the data will affect the effectiveness of machine learning.

[0244] In the field of machine learning, ground truth usually refers to data that is considered accurate or real.

[0245] Training datasets are used to train AI models. A training dataset can include the input to the AI ​​model, or it can include both the input and the target output of the AI ​​model. Specifically, a training dataset includes one or more training data points, which can include training samples input to the AI ​​model or the target output of the AI ​​model. The target output can also be referred to as a label, sample label, or labeled sample. The label is the ground truth value.

[0246] In the field of communications, training datasets can include simulation data collected through simulation platforms, experimental data collected from experimental scenarios, or measured data collected in actual communication networks. Because the geographical environment and channel conditions where the data is generated vary—for example, indoor / outdoor conditions, movement speed, frequency bands, or antenna configurations—the collected data can be categorized during acquisition. For instance, data with the same channel propagation environment and antenna configuration can be grouped together.

[0247] Model training essentially involves learning certain features from training data. In training AI models (such as neural network models), the goal is to make the model's output as close as possible to the desired predicted value. This is achieved by comparing the network's current predictions with the target value and updating the weight vector of each layer based on the difference. (Of course, there's usually an initialization process before the first update, where parameters are pre-configured for each layer.) For example, if the network's prediction is too high, the weight vector is adjusted to predict a lower value. This adjustment continues until the AI ​​model can predict the target value or a value very close to it. Therefore, it's necessary to predefine "how to compare the difference between the predicted and target values," which is the loss function or objective function. These are important equations used to measure the difference between the predicted and target values. Taking the loss function as an example, a higher output value (loss) indicates a greater difference. Therefore, training the AI ​​model becomes a process of minimizing this loss, making the loss function value less than a threshold, or making the loss function value meet the target requirements. For example, assuming the AI ​​model is a neural network, adjusting the model parameters of the neural network includes adjusting at least one of the following parameters: the number of layers of the neural network, the width, the weights of the neurons, or the parameters in the activation function of the neurons.

[0248] Inference data can be used as input to a trained AI model for inference. During the model's inference process, the inference data is input into the AI ​​model, and the corresponding output, which is the inference result, is obtained.

[0249] 7. AI Model Design:

[0250] The design of an AI model mainly includes the data collection phase (e.g., collecting training data and / or inference data), the model training phase, and the model inference phase. It can also further include the application of the inference results.

[0251] In the aforementioned data collection phase, the data source provides both training and inference data. In the model training phase, the AI ​​model is obtained by analyzing or training the training data provided by the data source. The AI ​​model represents the mapping relationship between the model's input and output. Learning the AI ​​model through model training nodes is equivalent to learning the mapping relationship between the model's input and output using the training data. In the model inference phase, the AI ​​model trained in the model training phase is used to perform inference based on the inference data provided by the data source, yielding the inference result. This phase can also be understood as: inputting inference data into the AI ​​model, obtaining the output through the AI ​​model, which is the inference result. This inference result can indicate the configuration parameters used (executed) by the execution object, and / or the operations performed by the execution object. In the inference result application phase, the inference result is published. For example, the inference result can be uniformly planned by the actor entity, which can send the inference result to one or more execution objects (e.g., network devices or terminal devices) for execution. Furthermore, the actor entity can also provide feedback on the model's performance to the data source, facilitating subsequent model updates and training.

[0252] It is understood that a communication system may include network elements with AI capabilities. The aforementioned AI model design-related steps can be performed by one or more network elements with AI capabilities. In one possible design, AI functions (such as AI modules or AI entities) can be configured within existing network elements in the communication system to implement AI-related operations, such as AI model training and / or inference. For example, the existing network element could be a network device or a terminal device. Alternatively, in another possible design, an independent network element can be introduced into the communication system to perform AI-related operations, such as training an AI model. This independent network element can be called an AI network element or an AI node, etc., and this application embodiment does not limit this name. Exemplarily, the AI ​​network element can be directly connected to the network device in the communication system, or it can be indirectly connected to the network device through a third-party network element. The third-party network element can be a core network element such as an authentication management function (AMF) network element, a user plane function (UPF) network element, OAM, a cloud server, or other network elements, without limitation. Exemplarily, this independent network element can be deployed on one or more of the following: the network device side, the terminal device side, or the core network side. Optionally, it can be deployed on a cloud server, OTT, or OAM. For example, an AI network element 140 is introduced into the communication system shown in Figure 1.

[0253] The training and inference processes of the AI ​​model in the embodiments of this application will be further illustrated below.

[0254] Training data used to train an AI model includes training samples and sample labels. For example, in an AI-based beam management scenario, the training samples are the CSI measurements corresponding to each reference signal resource in reference signal resource set B, and the sample labels are the CSI measurements corresponding to each reference signal resource in reference signal resource set A. Reference signal resource set B is a subset of reference signal resource set A, or the signal angle corresponding to each reference signal resource in reference signal resource set B is greater than the signal angle corresponding to each reference signal resource in reference signal resource set A. For example, each reference signal resource in reference signal resource set B is a wide beam, and each reference signal resource in reference signal resource set A is a narrow beam.

[0255] The specific training process is as follows: The model training node uses the measured CSI values ​​corresponding to each reference signal resource in the reference signal resource set B to determine the input data of the initial AI model, and inputs this input data into the initial AI model to obtain the inference result of the initial AI model. This input data includes the measured CSI values ​​corresponding to each reference signal resource in the reference signal resource set B, or it includes data obtained by performing interpolation, normalization, or filtering on the measured CSI values ​​corresponding to each reference signal resource in the reference signal resource set B. The inference result includes the predicted CSI values ​​corresponding to each reference signal resource in the reference signal resource set A. The model training node calculates the difference between the predicted and measured CSI values ​​corresponding to at least one reference signal resource in the reference signal resource set A (i.e., calculates the difference between the inference result of the initial AI model and the corresponding sample label), that is, calculates the value of the loss function. The model training node updates the parameters in the initial AI model according to the value of the loss function, minimizing the difference between the inference result obtained by the updated AI model and the corresponding sample label, i.e., minimizing the loss function. For example, the loss function can be the minimum mean square error (MSE) or cosine similarity, etc. Repeating the above steps yields an AI model that meets the target requirements. The training nodes for this model can be terminal devices, network devices, or other network elements with AI capabilities in a communication system.

[0256] In this application, prediction information (i.e., predicted values) refers to the prediction results directly output by the AI ​​model, or the results obtained after data processing of the prediction results directly output by the AI ​​model. A single piece of prediction information refers to the prediction results directly output by the AI ​​model in a single prediction process, or the results obtained after processing the prediction results. The AI ​​model can be deployed on a terminal device or on an OTT device on the terminal device side. When the AI ​​model is deployed on an OTT device, the terminal device can receive the prediction results output by the AI ​​model from the OTT device.

[0257] AI technology can be applied to beam management to reduce overhead. In AI model-based beam management, the AI ​​model can be a regression model or a classification model. The terminal device or network device determines the input data of the AI ​​model based on the CSI measurement value corresponding to each reference signal resource in the reference signal resource set B (Set B). In existing schemes, the number of reference signal resources in the reference signal resource set B used by the AI ​​model during the training or inference phase is usually fixed.

[0258] When the AI ​​model is a regression model, during the training phase, the input to the AI ​​model is input data A. The inference result obtained by inputting this input data A into the untrained AI model is the inference result A. The true value of the AI ​​model is the true value A, and the loss function of the AI ​​model is the loss function A. Input data A is determined based on the measured CSI values ​​corresponding to each reference signal resource in the reference signal resource set B. For example, input data A can be the measured CSI values ​​corresponding to each reference signal resource in the reference signal resource set B, or it can be data obtained after processing the measured CSI values ​​corresponding to each reference signal resource in the reference signal resource set B, such as through normalization or filtering. This filtering process includes removing one or more measured values ​​(e.g., one or more smaller measured values) from the measured CSI values ​​corresponding to each reference signal resource in the reference signal resource set B. The inference result A includes the predicted CSI values ​​corresponding to each reference signal resource in the reference signal resource set A (Set A). The true value A is determined based on the measured CSI values ​​corresponding to each reference signal resource in the reference signal resource set A. For example, the true value A is the measured CSI value corresponding to each reference signal resource in the reference signal resource set A, or the true value A is the data obtained after processing the measured CSI values ​​corresponding to each reference signal resource in the reference signal resource set A, such as by normalization. The loss function A is determined based on the inference result A and the true value A. For example, the loss function is the average of the differences between the measured and predicted channel state information values ​​corresponding to at least one reference signal resource in the reference signal resource set A, the MSE of the differences, or the cosine similarity of the differences. During the training process, the AI ​​model needs to be trained using multiple input data A, but the number of reference signal resources in the reference signal resource set B corresponding to these multiple input data A is the same. Based on the loss function A, one or more methods such as gradient descent, momentum method, and adaptive learning rate method are used to adjust the parameters in the untrained AI model, so that the difference between the inference result A obtained using the adjusted AI model and input data A and the true value A is minimized, that is, the loss function A is minimized, thereby obtaining a trained AI model. It is understood that "well trained" in this application may mean that the performance of the AI ​​model has met the requirements, for example, the loss function is less than the required threshold.

[0259] It should be understood that the methods used to adjust the parameters of an AI model, such as gradient descent, momentum, and adaptive learning rate, are similar to existing methods for adjusting AI model parameters. The following explanation uses gradient descent as an example to illustrate the method for adjusting AI model parameters to obtain a well-trained AI model.

[0260] For example, the steps for adjusting the parameters in an AI model using gradient descent on a terminal device or training device are as follows:

[0261] (1) Initialize the parameters of the AI ​​model, that is, determine at least one initial value for each parameter in at least one parameter of the initial AI model.

[0262] (2) Use the current parameters of the AI ​​model to process the current input of the AI ​​model and obtain the output data of the AI ​​model.

[0263] (3) Determine the value of the loss function based on the output data obtained in the previous step and the real data corresponding to the output data.

[0264] (4) Determine the gradient of the loss function based on the value of the loss function.

[0265] (5) Update the parameters of the AI ​​model based on the gradient of the loss function.

[0266] (6) Repeat steps (2)-(5) above until the loss function converges or the preset number of iterations is reached. The convergence of the loss function includes the loss function reaching its minimum value. The AI ​​model that converges when the loss function converges is a pre-trained AI model, or the AI ​​model that reaches the preset number of iterations is a pre-trained AI model.

[0267] During the training phase, the reference signal resources in reference signal resource set A are real resources (i.e., actually configured resources). Reference signal resource set B is a subset of reference signal resource set A. Alternatively, the signal angle corresponding to each reference signal resource in reference signal resource set B is greater than the signal angle corresponding to each reference signal resource in reference signal resource set A. For example, each reference signal resource in reference signal resource set B is a wide beam, and each reference signal resource in reference signal resource set A is a narrow beam.

[0268] During the inference phase, the reference signal resources in reference signal resource set A are either real resources (i.e., actually configured resources) or virtual resources (i.e., not actually configured resources). Reference signal resource set B is a subset of reference signal resource set A. Alternatively, the signal angle corresponding to each reference signal resource in reference signal resource set B is greater than the signal angle corresponding to each reference signal resource in reference signal resource set A. For example, each reference signal resource in reference signal resource set B is a wide beam, and each reference signal resource in reference signal resource set A is a narrow beam.

[0269] When the AI ​​model is a regression model, during the inference phase, the input to the AI ​​model is input data B. The inference result obtained by inputting this input data B into the trained AI model is the inference result B. This input data B is determined based on the measured CSI values ​​corresponding to each reference signal resource in the reference signal resource set B. For example, input data B can be the measured CSI values ​​corresponding to each reference signal resource in the reference signal resource set B, or it can be data obtained after processing the measured CSI values ​​corresponding to each reference signal resource in the reference signal resource set B, such as through normalization or filtering. The inference result B includes the predicted CSI values ​​corresponding to each reference signal resource in the reference signal resource set A. Based on the inference result B, the terminal device determines at least one optimal reference signal resource predicted in the entire set or subset M of the reference signal resource set A (Set A), and reports it to the network device, for example, reporting the identification information (ID) of the at least one optimal reference signal resource predicted and / or the predicted channel state information corresponding to the at least one optimal reference signal resource predicted. The predicted channel state information (CSO) value corresponding to at least one optimal reference signal resource in reference signal resource set A is greater than or equal to the predicted CSO values ​​corresponding to all reference signal resources in reference signal resource set A except for the at least one optimal reference signal resource. The predicted CSO value corresponding to at least one optimal reference signal resource in subset M is greater than or equal to the predicted CSO value corresponding to all reference signal resources in subset M except for the at least one optimal reference signal resource, or the at least one optimal reference signal resource in subset M is a reference signal resource in subset M that belongs to the at least one optimal reference signal resource in reference signal resource set A. The reference signal resources in reference signal resource set B used in the training and inference phases of the AI ​​model can be the same or different. However, when the number of reference signal resources in reference signal resource set B used in the inference phase of the AI ​​model differs from the number of reference signal resources in reference signal resource set B used in the training phase of the AI ​​model, the accuracy of the inference result obtained by the AI ​​model in the inference phase is lower.

[0270] When the AI ​​model is a classification model, during the training phase, the input to the AI ​​model is input data C. The inference result obtained by inputting this input data C into the untrained AI model is the inference result C. The true value of the AI ​​model is the true value C, and the loss function of the AI ​​model is the loss function C. The input data C is determined based on the measured values ​​of the channel state information corresponding to each reference signal resource in the reference signal resource set B. For example, the input data C can be the measured values ​​of the channel state information corresponding to each reference signal resource in the reference signal resource set B, or it can be data obtained after processing the measured values ​​of the channel state information corresponding to each reference signal resource in the reference signal resource set B, such as through normalization or filtering. The inference result C is determined based on the probability that each reference signal resource in the reference signal resource set A is the best reference signal resource in the reference signal resource set A. For example, the inference result C includes the probability that each reference signal resource in the reference signal resource set A is the best reference signal resource in the reference signal resource set A, or it can include the predicted k1 best reference signal resources in the reference signal resource set A. The measured value of the channel state information corresponding to the best reference signal resource in the reference signal resource set A is greater than or equal to the measured values ​​of the channel state information corresponding to other reference signal resources in the reference signal resource set A besides the best reference signal resource. That is, the best reference signal resource in the reference signal resource set A is the reference signal resource with the best signal transmission performance in the reference signal resource set A. The probability corresponding to the predicted k1 best reference signal resources is greater than or equal to the probability corresponding to the reference signal resources in the reference signal resource set A besides the predicted k1 best reference signal resources. The probability corresponding to the reference signal resource is the probability that the reference signal resource is the best reference signal resource in the reference signal resource set A. The truth value C is determined based on the measured value of the channel state information corresponding to each reference signal resource in the reference signal resource set A. For example, the truth value C includes the measured k1 best reference signal resources in the reference signal resource set A. The measured value of the channel state information corresponding to the measured k1 best reference signal resources is greater than or equal to the measured value of the channel state information corresponding to the reference signal resources in the reference signal resource set A besides the measured k1 best reference signal resources. The loss function C is determined based on the inference result C and the true value C. For example, the loss function C is the number of different reference signal resources among the predicted and measured k1 optimal reference signal resources in the reference signal resource set A, or the loss function C is the ratio of the number of different reference signal resources among the predicted and measured k1 optimal reference signal resources in the reference signal resource set A to k1. Here, k1 is a positive integer.During training, an AI model requires multiple input data sets C, but the number of reference signal resources in the corresponding reference signal resource set B is the same. Based on the loss function C, one or more methods, such as gradient descent, momentum, and adaptive learning rate, are used to adjust the parameters of the untrained AI model. This minimizes the difference between the inference result C obtained using the adjusted AI model and the input data C and the true value C, i.e., minimizing the loss function C, thereby obtaining a well-trained AI model.

[0271] When the AI ​​model is a classification model, during the inference phase, the input to the AI ​​model is input data D. The inference result obtained by inputting this input data D into the trained AI model is the inference result D. This input data D is determined based on the measured values ​​of the channel state information corresponding to each reference signal resource in the reference signal resource set B. For example, input data D can be the measured values ​​of the channel state information corresponding to each reference signal resource in the reference signal resource set B, or it can be data obtained after processing the measured values ​​of the channel state information corresponding to each reference signal resource in the reference signal resource set B, such as through normalization or filtering. The inference result D includes the probability that each reference signal resource in the reference signal resource set A is the best reference signal resource in that set A. Based on the inference result D, the terminal device determines at least one predicted best reference signal resource in the entire set or subset M of the reference signal resource set A and reports it to the network device, for example, reporting the identification information of the predicted at least one best reference signal resource. The probability corresponding to at least one optimal reference signal resource determined by prediction in subset M is greater than or equal to the probability corresponding to any other reference signal resource in subset M besides the at least one optimal reference signal resource determined by prediction; or, the at least one optimal reference signal resource determined by prediction in subset M is a reference signal resource in subset M that belongs to at least one optimal reference signal resource determined by prediction in reference signal resource set A. The reference signal resources in reference signal resource set B used in the training and inference phases of the AI ​​model can be the same or different.

[0272] For example, when applying AI in beam management scenarios, terminal devices or network devices can efficiently and accurately identify the optimal beam using AI models. This AI model can be located in either the terminal device or the network device.

[0273] For ease of understanding and explanation, beam and reference signal resource will be described as two separate concepts in the following text. Reference signal resource can include time-frequency resources used for transmitting reference signals, which can be indicated through reference signal resource configuration. A beam can refer to the spatial resources of the reference signal transmitted through the reference signal resource. It can be understood that each reference signal resource can correspond to one beam, and each set of reference signal resources can correspond to a set of beams. Reference signals transmitted through different reference signal resources can correspond to the same spatial resource, i.e., the same beam; or they can correspond to different spatial resources, i.e., different beams.

[0274] Specifically, AI-based beam management includes two typical use cases: spatial beam prediction and temporal beam prediction.

[0275] A use case for spatial beam prediction is shown in Figure 6. Traditional beam measurement requires scanning all beams in the selectable beam set using a P1-P3 process to obtain the optimal (e.g., the beam with the highest RSRP) beam. For systems using massive MIMO antennas, the selectable beam set can be very large (e.g., 1024 beams), and the traditional beam scanning process incurs significant measurement overhead. With the introduction of AI, only a portion of the selectable beam set (e.g., Set A, the set of boxes with and without filled patterns in the figure) can be measured (e.g., Set B, the set of boxes with filled patterns in the figure). Based on the measurements of the portion of the beams, the top K beams in the full beam set (Set A) can be predicted, thus significantly reducing beam measurement overhead.

[0276] The use case of time-domain beam prediction is shown in Figure 7. The AI ​​beam prediction model can use historical beam measurement information to predict future beam information (top K beam indices, corresponding RSRPs), thereby improving beam management robustness in scenarios with rapidly changing channels and avoiding frequent beam measurements and handovers. The beam prediction model can be located in terminal devices or network devices.

[0277] The lifecycle of an AI model involves the following stages: data collection, model training (or model learning), model information dissemination, model activation / deactivation, model inference (or model reasoning, inference, or prediction), model monitoring or validation, model updates, or inference result dissemination. Air interface AI model lifecycle management (LCM) can be based on either model ID or functionality. A model ID is an identifier assigned in some way to identify the model. In model ID-based LCM, the model ID indicates operations performed on the model, such as activation / deactivation / selection / rollback / switching. AI features refer to the characteristics that allow the use of AI, or features that require the use of the AI ​​model, such as AI-based CSI feedback or AI-based beam management. Functions refer to configuration-related AI features, where the configuration is supported by conditions indicated by the terminal device's capabilities. In other words, a function corresponds to a specific configuration under an AI feature, such as AI-based CSI feedback or AI-based beam management under a specific configuration. In function-based LCM, network devices can instruct terminal devices to operate AI functions, such as activation / deactivation / selection / rollback / switching, through 3GPP signaling (e.g., RRC signaling, MAC-CE, DCI).

[0278] Specifically, an AI feature may encompass one or more functions. For each function, there may be one or more models used to implement it. Conversely, a single model can also be used to implement multiple functions. The following examples illustrate the relationship between functions and AI features:

[0279] Example 1: One function corresponds to one AI feature. For example, function 1 is AI-based temporal beam prediction, and function 2 is AI-based spatial beam prediction.

[0280] Example 2: One function corresponds to one AI feature + a set of specific RRC configurations. For example: Function (1) is AI-based time-domain beam prediction under configuration 1, and Function (2) is AI-based time-domain beam prediction under configuration 2. The configurations here include: CSI-RS resource configuration, CSI reporting configuration, beam set configuration, prediction window configuration, etc.

[0281] Example 3: One function corresponds to one AI feature + a specific set of RRC configurations + scenario / site identifier. For example, function 1 is AI-based temporal beam prediction under configuration 1 and scenario 1, and function 2 is AI-based temporal beam prediction under configuration 1 and scenario 2. The configurations here include: CSI-RS resource configuration, CSI reporting configuration, beam set configuration, and prediction window configuration. Scenarios can be: urban areas, suburbs, urban macrocells (UMa), urban microcells (UMi), indoor hotspot cells (InH), highways, etc.

[0282] In some scenarios, such as interference-cooperative scenarios, network devices may only use a subset of beams in Set A for transmission within a specific cell. Therefore, network devices can only transmit signals using a subset of beams from Set A in a given cell to avoid interference with neighboring cells. Consequently, network devices prefer to obtain information about a subset of beams in Set A (i.e., subSet A) rather than focusing on the overall beam performance of Set A.

[0283] In other scenarios, such as inter-cell interference cooperation, the top K beams in a subset of Set A may not perform well within the overall Set A, while better beams exist outside the subset. Therefore, network devices can subsequently adjust the beams to make better beams available within the cell. Consequently, network devices may want to obtain beam information for both Set A and a subset of beams within Set A (e.g., a subset of Set A).

[0284] To address the different needs mentioned above, network devices can use different reporting configurations to instruct terminal devices to report for Set A and / or subsets of Set A respectively, which may result in significant configuration overhead.

[0285] In view of this, this application provides a method to reduce configuration overhead by associating a reporting configuration with multiple resource configurations, which may include configurations of reference signal resource sets corresponding to the full beam set and subsets of the full beam set, respectively.

[0286] For ease of understanding and explanation, in the embodiments shown below, Set A and Set B are used to represent different beam sets. Set A can be considered a full beam set, and Set B may or may not be a subset of Set A. For example, the beams in Set B may be wide beams, while the beams in Set A may be narrow beams. Based on the relationship between reference signal resources and beams described above, Set A and Set B can also correspond to different sets of reference signal resources. Therefore, in some other implementations, Set A and Set B can also be used to represent different sets of reference signal resources, corresponding to different beam sets.

[0287] The method provided in this application will now be described in detail with reference to the accompanying drawings.

[0288] It should be noted that the first or third device in the following embodiments can be replaced by a device on the terminal device side, and the second or fourth device can be replaced by a device on the network device side or a device on the core network (CN) side.

[0289] Devices on the terminal device side can include the terminal device itself, the communication module within the terminal device, or the circuits or chips within the terminal device responsible for communication functions (such as a modem chip, also known as a baseband chip, or a SoC chip or SIP chip containing a modem core, etc.), or can be included in AI entities on the terminal device side. AI entities on the terminal device side can be the terminal device itself or AI entities serving the terminal device, such as servers, like OTT servers or cloud servers. Devices on the network device side can include the network device itself, the communication module within the network device, or the circuits or chips within the network device responsible for communication functions (such as a modem chip, also known as a baseband chip, or a SoC chip or SIP chip containing a modem core, etc.), or can include AI entities on the network device side. AI entities on the network device side can be the network device itself or AI entities serving the network device, such as RIC, OAM, or servers, like OTT servers or cloud servers. The equipment on the core network side can include the core network elements themselves, communication modules within the core network elements, or circuits or chips within the core network elements responsible for communication functions (such as modem chips, also known as baseband chips, or system-on-a-chip (SoC) chips or SIP chips containing modem cores, etc.), or it can include AI entities on the core network side. AI entities on the core network side can be the core network elements themselves, or they can be AI entities serving the core network elements, such as servers, like OTT servers or cloud servers.

[0290] Figure 8 is a schematic flowchart of a communication method 800 provided in an embodiment of this application. For ease of understanding and explanation, the method shown in Figure 8 uses a terminal device as an example of a first device and a network device as an example of a second device, illustrating the implementation flow of the method 800.

[0291] As shown in Figure 8, the method includes steps 810 and 820. Optionally, it also includes one or more of steps 830 to 850. The various steps in method 800 are described in detail below.

[0292] In step 810, the network device sends a reported configuration, which is associated with multiple resource configurations, including a first resource configuration, P second resource configurations, and a third resource configuration. Correspondingly, the terminal device receives the reported configuration.

[0293] Each of these multiple resource configurations can be used to configure a set of reference signal resources. Each set of reference signal resources can correspond to a beam set. Each set of reference signal resources includes one or more reference signal resources, and each reference signal resource corresponds to a beam in the beam set. Therefore, it can be said that each resource configuration corresponds to a beam set.

[0294] For ease of distinction and explanation, the beam set corresponding to the reference signal resource set configured by the first resource configuration is referred to as the first beam set, the beam set corresponding to the reference signal resource set configured by the second resource configuration is referred to as the second beam set, and the beam set corresponding to the reference signal resource set configured by the third resource configuration is referred to as the third beam set.

[0295] Optionally, P is 1. That is, the reported configuration is associated with three resource configurations: the first resource configuration, the second resource configuration, and the third resource configuration.

[0296] In this embodiment, the P second resource configurations and the first resource configuration satisfy the following condition: the beam set corresponding to the reference signal resource set configured in each of the P second resource configurations is a subset of the beam set corresponding to the reference signal resource set configured in the first resource configuration. That is, each of the P second beam sets is a subset of the first beam set. For ease of understanding and explanation, the relationship satisfied between the P second resource configurations and the first resource configuration is referred to as the first relationship.

[0297] The beam set corresponding to the reference signal resources configured in the third resource configuration can be a subset of the beam set corresponding to the reference signal resources configured in the first resource configuration; that is, the third beam set can be a subset of the first beam set. Alternatively, the signal angle of the beams corresponding to the reference signal resources configured in the third resource configuration is greater than the signal angle of the beams corresponding to the reference signal resources configured in the first resource configuration; that is, the third beam set does not have to be a subset of the first beam set. For example, the beams in the third beam set are wide beams, and the beams in the first beam set are narrow beams.

[0298] As discussed above regarding the application of AI in beam management, an AI model can predict the top K beams in the full beam set (e.g., Set A) based on measurement results of a subset of beams (e.g., Set B). In this embodiment, the first beam set corresponding to the first resource configuration can be considered as Set A, and the beam set corresponding to the third resource set can be considered as Set B. The second beam set corresponding to each of the P second resource sets is a subset of Set A.

[0299] For example, the reporting configuration may be a CSI reporting configuration (CSI-ReportConfig) carried by an RRC message, and the multiple resource configurations may be CSI resource configurations (CSI-ResourceConfig) associated with CSI-ReportConfig. The set of reference signal resources configured by each resource configuration may be a CSI-RS resource Set configured by CSI-ResourceConfig.

[0300] The reporting configuration is associated with multiple resource configurations. This can be understood as the reporting configuration containing multiple resource configurations, or the reporting configuration referencing multiple resource configurations, or multiple resource configurations being used for the reporting configuration.

[0301] In this application, "report configuration" can also be replaced with "report configuration".

[0302] In this embodiment, there are multiple ways to report configurations associated with multiple resources. Two possible configuration methods are illustrated below.

[0303] Configuration Method 1: In the reported configuration, simultaneously associate the first resource configuration, P second resource configurations, and the third resource configuration.

[0304] For example, the reporting configuration may include an identifier for a first resource configuration, identifiers for P second resource configurations, and an identifier for a third resource configuration, associating the first resource configuration with the P second resource configurations. Based on this reporting configuration, the terminal device can determine a first task: performing a measurement based on the set of reference signal resources configured by the third resource configuration to obtain a measurement result; and performing inference based on the measurement result to obtain an inference result for the first resource configuration and the set of reference signal resources configured by the P associated second resource configurations. In other words, the first task may be to predict and report the set of reference signal resources configured for the first resource configuration and the P associated second resource configurations.

[0305] For example, the reporting configuration is CSI-ReportConfig. This CSI-ReportConfig can simultaneously associate Set A, a subset of Set A, and Set B. The reporting behavior of the terminal device can be specified by the protocol or instructed by the network device. Specifically, the reporting behavior of the terminal device can be based on the measurement results of Set B, and predictions can be made and reported for both Set A and a subset of Set A. Therefore, this reporting configuration can be regarded as an enhancement of the current CSI-ReportConfig.

[0306] Configuration Method 2: Configure multiple sub-reporting configurations (or sub-reporting configurations) in the reporting configuration, with each sub-reporting configuration associated with a resource configuration.

[0307] For example, the reporting configuration may include an identifier for a first resource configuration, an identifier for a third resource configuration, and multiple sub-reporting configurations. Each sub-reporting configuration contains a list of reference signal resources for that sub-reporting configuration. This list of reference signal resources (e.g., an NZP-CSI-RS resource list (NZP-CSI-RS-ResourceList)) indicates a set of reference signal resources. The set of reference signal resources indicated by each sub-reporting configuration may be a subset of the set of reference signal resources configured by the first resource configuration. Specifically, the values ​​in the reference signal resource list are non-negative integers, where a value of 0 represents the first NZP CSI RS resource in the set of reference signal resources configured by the first resource configuration, a value of 1 represents the second NZP CSI RS resource in the set of reference signal resources configured by the first resource configuration, and so on, with the maximum value being maxNrofNZP-CSI-RS-ResourcesPerSet-1, where maxNrofNZP-CSI-RS-ResourcesPerSet is the total number of reference signal resources in the set of reference signal resources configured by the first resource configuration. By including a list of reference signal resources in each sub-reporting configuration, each sub-reporting configuration can be associated with a resource configuration. In this embodiment, the multiple sub-reporting configurations can specifically be (P+1) sub-reporting configurations, each associated with a first resource configuration and P second resource configurations respectively.

[0308] For example, the reporting configuration is CSI-ReportConfig, which includes multiple CSI-ReportSubConfigs. One CSI-ReportSubConfig is associated with Set A (the nzp-CSI-RS-ResourceList-r18 field takes values ​​from 1 to maxNrofNZP-CSI-RS-ResourcesPerSet-1, representing all reference signal resources in the reference signal resource set configured in the first resource configuration). Another one or more CSI-ReportSubConfigs are associated with one or more subsets of Set A (the nzp-CSI-RS-ResourceList-r18 field takes values ​​from 1 to maxNrofNZP-CSI-RS-ResourcesPerSet-1, representing some reference signal resources in the reference signal resource set configured in the first resource configuration). Here, r18 represents the protocol version 18 (release 18). The reporting behavior of the terminal device can be specified by the protocol or instructed by the network device. Specifically, the terminal device's reporting behavior can be based on the measurement results of Set B, predicting and reporting for both Set A and subsets of Set A.

[0309] Figure 9 shows an example of a reporting configuration. Figure 9 illustrates an example where P=1. As shown, the reporting configuration is CSI-ReportConfig, and multiple sub-reporting configurations are shown as CSI-ReportSubConfig1 and CSI-ReportSubConfig2. CSI-ReportConfig contains configuration information for Set A and Set B, where Set A includes 128 beams. CSI-ReportSubConfig1 contains configuration information for Set A, indicating that Set B should be measured and Set A predicted. CSI-ReportSubConfig2 contains configuration information for a subset of Set A, indicating that Set B should be measured and a subset of Set A predicted, which includes 64 beams. It should be understood that Figure 9 is merely an example and should not constitute any limitation on the number of second resource configurations or the number of beams contained in the beam set corresponding to the reference signal resource set configured in the first resource configuration.

[0310] As mentioned earlier, if a CSI-RS resource is referenced N times by one or more CSI reporting configurations, the count is N. In this embodiment, since the network device associates a first resource configuration, P second resource configurations, and a third resource configuration through the reporting configuration, that is, Set B is referenced once through one CSI reporting configuration, the count can be 1. This avoids duplicate counting of Set B resources and prevents duplicate CPU usage. In addition, associating multiple resource configurations through the reporting configuration can also save configuration overhead.

[0311] In step 820, the terminal device determines first inference information based on the reported configuration. The first inference information indicates the inference results of the reference signal resource sets configured for the first resource configuration and P second resource configurations, respectively.

[0312] The terminal device can determine the reference signal resource sets associated with each of the multiple resource configurations reported, and then, based on the reporting behavior of the terminal device as specified in the protocol or instructed by the network device, obtain first inference information through model inference. This first inference information can indicate the inference results for the reference signal resource sets configured for the first resource configuration and P second resource configurations respectively; that is, the inference results indicate the inference results for the reference signal resource sets configured for the first resource configuration, and the inference results for the reference signal resource sets configured for each of the P second resource configurations. The aforementioned inference results can be obtained based on measurement results obtained by measuring the reference signal resource sets configured for the third resource configuration.

[0313] It is understandable that the model can be deployed on a terminal device, thus allowing the terminal device to perform model inference independently and obtain first inference information, and optionally, determine the second inference information described later. Alternatively, the model can be deployed on other devices besides the terminal device, such as inference nodes. In this case, the terminal device can be called a data collection node. The data collection node can collect inference results from the inference node and then determine the first inference information based on the inference results, and optionally, determine the second inference information as well. In other words, the data collection node and the inference node can be the same device or different devices.

[0314] In this embodiment, the inference result for the reference signal resource set configured for the first resource configuration and the inference result for the reference signal resource sets configured for P second resource configurations can be obtained by the same model, which can be trained using method 1400 described below. In other words, the model has the ability to determine the inference result for multiple reference signal resource sets with a first relationship, or the model has the ability to determine the inference result for reference signal resource sets configured for the first resource configuration and the P second resource configurations respectively. Since the model can obtain the inference result by performing an inference task (e.g., denoted as the first task), it can also be said that the first task has the ability to determine the inference result for multiple reference signal resource sets with a first relationship, or the first task has the ability to determine the inference result for reference signal resource sets configured for the first resource configuration and the P second resource configurations respectively. Since the model can be deployed on terminal devices or on devices other than terminal devices (such as OTT servers), it can also be said that the device with the model deployed (such as a terminal device or an OTT server) has the ability to support determining the inference result for multiple reference signal resource sets with a first relationship. In other words, the device with the model deployed (such as a terminal device or an OTT server) has the ability to support determining the inference result for reference signal resource sets configured with a first resource configuration and P second resource configurations respectively.

[0315] The terminal device can obtain measurement results based on the measurement of the reference signal resource set configured by the third resource configuration, and then use the measurement results as input data for the model, so as to perform the first task and obtain the first inference information through the model.

[0316] As mentioned earlier, AI can be applied to beam management. Therefore, the first inference information obtained by the model in performing the first task can be information about one or more beams. These one or more beams can be the top-ranked beams in a beam set based on their measurement values. Measurement values ​​can include, but are not limited to, RSRP, RSRQ, SINR, etc. For ease of explanation, the top-ranked beams will be referred to as "top beams" below. For example, "top K beams" represents the top K beams based on their measurement values. It can be understood that the top K beams are the K beams with the largest measurement values, or the K beams with the best quality. Beam information can include beam identifiers, beam measurement values, etc. The information reported for each beam can be indicated by the reporting configuration or predefined by the protocol; this application does not limit this.

[0317] The following examples illustrate three possible designs for the content contained in the first inference information.

[0318] In a first possible design, the first inference information includes information about multiple beams, which is the complete set of the following terms: the top K beams in the first beam set, and the top K beams in the p-th second beam set obtained by iterating through p from 1 to P. p The beams, the first beam set corresponds to the reference signal resource set configured in the first resource configuration, the p-th second beam set corresponds to the reference signal resource set configured in the p-th second resource configuration among the P second resource configurations, p, K and K p It is a positive integer.

[0319] The above K, K1 to K P One or more of these can be indicated by the network device, for example, in the reporting configuration. Accordingly, the reporting configuration also indicates one or more of the following: K, K1 to K P Or, K, K1 to K P One or more of these can be predefined by the protocol. In other words, K, K1 to K... P This can be indicated by the network device and / or predefined by the protocol. For example, the protocol predefines some items, and the network device indicates others; or, the protocol predefines all items; or, the network device indicates all items. In this case, the number of top beams inferred for each beam set can be predetermined.

[0320] That is, the aforementioned multiple beams can be the complete set of the following terms: the top K beams in the first beam set, the top beams in each of the P second beam sets, that is, {the top K beams in the first beam set, the top K1 beams in the 1st second beam set, the top K2 beams in the 2nd second beam set, ..., the top K beams in the Pth second beam set} P Beam.

[0321] Since each of the P second beam sets is a subset of the first beam set, one or more of the top beams in the P second beam sets may partially or completely overlap with the top K beams in the first beam set. Furthermore, two or more of the top beams in the P second beam sets may also partially or completely overlap. Therefore, one possible implementation of the complete set of the above terms is the result after deduplicating the top beams of the terms. Deduplication treats overlapping beams as a single beam. Deduplication reduces reporting overhead. In this case, the number of top beams reported for the first beam set and the P second beam sets may not be fixed, and the terminal device can additionally indicate the number of top beams corresponding to the beam information reported for the first beam set and the P second beam sets. Another possible implementation of the complete set of the above terms is to indicate the top beams of each term separately without deduplication, thereby fixing the number of beams reported for each reference signal resource set and fixing the number of reported bits.

[0322] In the second possible design, the first inference information includes: a first inference result and P second inference results; wherein, the p-th second inference result among the P second inference results includes K from the p-th second beam set among the P second beam sets. p Information on each beam; the first inference result includes information on K' beams in the first beam set, where K' beams are the K' beams in the first beam set excluding K” beams, and K” beams are the complete set of P second beam sets; K', K”, K p Both p and p are positive integers, with p taking values ​​from 1 to p.

[0323] The above K', K1 to K P One or more of these can be indicated by the network device, for example, in the reporting configuration. Accordingly, the reporting configuration also indicates one or more of the following: K', K1 to K P Or, K', K1 to K P One or more of these can be predefined by the protocol. In other words, K', K1 to K' above... P This can be indicated by the network device and / or predefined by the protocol. For example, the protocol predefines some items, and the network device indicates others; or, the protocol predefines all items; or, the network device indicates all items. In this case, the number of beams in the first inference result and P second inference results can be predetermined. That is, the number of beams corresponding to the beam information in the first inference information can be predetermined, and the beam information in the first inference result and the beam information in the P second inference results do not overlap. More useful beam information can be reported under the same reporting overhead.

[0324] The K' beams in the first beam set can be the top K' beams excluding the entire set of P second beam sets in the first beam set. That is, the beam information in the first inference information that overlaps with the top beams inferred from the first beam set and the P second beam sets will not be reported repeatedly.

[0325] Optionally, in the second possible design described above, the P sets of second beams can be different from each other, that is, there are no overlapping beams in the top beams inferred from the P sets of second beams.

[0326] Optionally, in the second possible design, P=1, that is, the first inference information includes: the top K1 beam in the second beam set (a subset of Set A), and the top K' beam in the beams in the first beam set other than the second beam set (beams in Set A other than a subset of Set A).

[0327] In the third possible design, the first inference information includes: a first inference result and P second inference results; wherein, the first inference information includes K from the first beam set. sum Information about each beam; the first inference result includes (K) beams from the first beam set excluding the K”' beams. sum -K”') beam information; K”' beams are the K beams in the p-th second beam set obtained by iterating through p from 1 to P. p The complete set of P beams, where the p-th second inference result in the P second inference results includes K in the p-th second beam set of the P second beam sets. p Information about each beam, K p Greater than or equal to K p,min K p,min Less than K sum ,K”'、K sum K p K p,min Both p and p are positive integers, with p taking values ​​from 1 to p.

[0328] In other words, the first inference information includes K in the first beam set. sum Information about each beam, the K sum Each beam includes K from the p-th second beam set obtained by iterating through p from 1 to P. p The complete set of beams, i.e., K”' beams, and the set of beams in the first beam set excluding the K”' beams (K sum -K”') beams.

[0329] The above K sum K 1,min To KP,min One or more of these can be indicated by the network device, for example, in the reporting configuration. Accordingly, the reporting configuration also indicates one or more of the following: K sum K 1,min To K P,min Or, K sum K 1,min To K P,min One or more of the terms can be predefined by the protocol. In other words, the above K... sum K 1,min To K P,min This can be indicated by the network device and / or predefined by the protocol. For example, the protocol predefines some items, and the network device indicates others; or, the protocol predefines all items; or, the network device indicates all items. In this case, the total number of top beams inferred for all beam sets can be predetermined, and the top beams of the first beam set and the second beam set do not overlap.

[0330] Here, K”' beams are the complete set of the top beams of the P second beam sets. The number of top beams inferred for the p-th second beam set is greater than or equal to K. p,min That is, not less than K p,min That is, for the p-th second beam set, at least K... p,min There are P top beams. Since each of these P second beam sets is a subset of the first beam set, there may be two or more second beam sets where the top beams partially or completely overlap. Therefore, one possible implementation of the complete set of top beams for the P second beam sets is the result after deduplicating the top beams of the P second beam sets. Deduplication treats overlapping beams as a single beam. Deduplication reduces unnecessary reporting overhead. The complete set of top beams that can be determined for these P second beam sets can be at most [number missing]. There are P beams, and considering that the top beam in this set of second beams may overlap, therefore... Another possible implementation of the complete set of the top beams of the aforementioned P second beam sets is to indicate the top beams of each of the P second beam sets separately without deduplication. This allows us to fix the number of beams reported for each reference signal resource set, and thus the number of bits reported. Given a fixed total number of top beams inferred for all beam sets, the number of top beams inferred for the first beam set can be calculated as (K). sum-K”'). The top beam for inference of the first beam set can be the top beam of the first beam set excluding the aforementioned K”' beams. That is, the top beam obtained by the first inference information for the first beam set does not overlap with the top beam obtained by inference of the P second beam sets.

[0331] Optionally, in the third possible design, P = 1, that is, the first reasoning information includes K. sum A fixed number of beam information, including at least the top K beams from the second beam set (Set A subset). 1,min K1 beams. That is, if the first inference information contains K1 top beams from the second beam set (Set A subset), then the remaining (K1) top beams in the first inference information... sum -K1) beams are the top (K1) beams in the first beam set excluding the top K1 beams in the second beam set (beams in Set A excluding subsets of Set A). sum -K1) beam.

[0332] Understandably, in the second and third possible designs described above, since the P sets of second beams are subsets of the first beam set, and the P sets of second beams can be distinct from each other, meaning that the top beams inferred from these P sets of second beams may not have overlapping beams, the complete set of top beams inferred from these P sets of second beams can be... A single beam can be used without performing a deduplication operation.

[0333] For example, the first beam set includes multiple beams: {#1, #2, #3, #4, #5, #6, #7, #8}. The second beam set is a subset of the first beam set. By reasoning, the beam quality in the first beam set, from highest to lowest, is {#1, #2, #3, #4, #5, #6}. In scenario one, the second beam set includes beams {#2, #6, #7}, and in scenario two, the second beam set includes beams {#1, #2, #3, #8}. See Table 1 below for details.

[0334] Table 1

[0335] Table 1 shows the top beams determined from the second beam set in both Scenario 1 and Scenario 2. Specifically, the top beams determined from the second beam set in Scenario 1 include {#2, #6}, and the top beams determined from the second beam set in Scenario 2 include {#1, #2, #3}. Based on the three different designs provided above, the first inference information is shown in Table 2.

[0336] Table 2

[0337] It is easy to see that in the three different designs mentioned above, when the top beam of the second beam set has a high degree of overlap with the top beam of the first beam set, the third design can report more of the top beams in the second beam set (that is, a subset of the first beam set).

[0338] It should be understood that the examples in Tables 1 and 2 above are provided for ease of understanding only and should not constitute any limitation on the number of beams and beam identifiers included in the first beam set, the second beam set, or the number of top beams and beam identifiers in each beam set.

[0339] In step 830, the terminal device sends first inference information. Correspondingly, the network device receives the first inference information.

[0340] The terminal device can report the first inference information to the network device. For example, the terminal device can send a CSI report to the network device, which carries the first inference information. This CSI report can be generated based on a CSI-ReportConfig (i.e., an example of a reporting configuration) sent by the network device.

[0341] The terminal device does not necessarily send the first inference information every time it receives it. In the event of insufficient resources, the terminal device may discard the received first inference information or wait until resources are sufficient before sending it; this application does not limit this. In other words, step 830 is an optional step and does not necessarily have to be performed.

[0342] Optionally, the terminal device may periodically send the first inference information.

[0343] Step 830 may specifically include: the terminal device sending first inference information at a first reporting time, the first reporting time corresponding to a first reporting period. Correspondingly, the network device receiving the first inference information at the first reporting time.

[0344] The first reporting period can refer to the reporting period of the first inference information. The first reporting time corresponds to the first reporting period; that is, the first reporting time is determined based on the first reporting period. The time offset between two adjacent first reporting times can be considered as one first reporting period. After determining the first first reporting time, the terminal device can periodically report the first inference information using the first reporting period as the cycle.

[0345] Optionally, the first reporting period can be indicated by the reporting configuration; in other words, the reporting configuration also indicates the first reporting period. Optionally, the first reporting period is predefined by the protocol.

[0346] Since the first inference information includes the inference results for the first resource configuration and the inference results for P second resource configurations, it can be assumed that the reporting period configured for the first resource configuration and the P second resource configurations are the same, or the reporting period configured for the first resource configuration can be an integer multiple of the reporting period configured for the P second resource configurations. This can save some reporting overhead. For example, in the aforementioned configuration method one, a reporting period, namely the first reporting period, can be configured through reporting configuration. For example, in the aforementioned configuration method two, a reporting period can be configured for the first resource configuration and P second resource configurations, i.e., the first reporting period, which can be indicated in CSI-ReportConfig; or, a reporting period can be configured separately for each of the first resource configuration and each of the P second resource configurations, thus configuring (P+1) reporting periods. For example, the first reporting period and P second reporting periods can be indicated in CSI-ReportConfig, or the first reporting period can be indicated in CSI-ReportConfig, and the second reporting period can be indicated in each CSI-ReportSubConfig. The reporting period configured for the first resource configuration is the first reporting period, and the reporting period configured for the p-th second resource configuration among the P second resource configurations is the p-th second reporting period among the P second reporting periods. The P second reporting periods configured for the P second resource configurations can be the same or different. The first reporting period can be an integer multiple of any of the P second reporting periods; that is, the first reporting period can be a common multiple of the P second reporting periods.

[0347] Optionally, the method further includes step 840: the terminal device sends second inference information at the second reporting time, the second inference information including inference results for reference signal resource sets configured for one or more of the P second resource configurations respectively. Accordingly, the network device receives the second inference information at the second reporting time.

[0348] In other words, given sufficient resources, the terminal device can send second inference information in addition to the first inference information during the first reporting cycle, and also send second inference information during the second reporting cycle. For example, the second inference information can also be included in the CSI report.

[0349] In this embodiment, the second reporting time corresponds to the second reporting period; that is, the second reporting time can be determined based on the second reporting period. Since a second reporting period can be configured for each of the P second resource configurations, and these P second reporting periods can be the same or different, the inference result of the reference signal resource set configured for the p-th second resource configuration is sent at the p-th second reporting time, which corresponds to the p-th second reporting period.

[0350] Optionally, the P second reporting cycles can be indicated by the reporting configuration; in other words, the reporting configuration also indicates the P second reporting cycles. Optionally, the P second reporting cycles are predefined by the protocol.

[0351] Figure 10 is a schematic diagram illustrating the transmission of first inference information at a first reporting time and the transmission of second inference information at a second reporting time, according to an embodiment of this application.

[0352] As shown in Figure 10(a), the first reporting period is the same as P second reporting periods, or in other words, the network device configures the same reporting period for the first resource configuration and P second resource configurations, i.e., the first reporting period. Sets #1 and #2 in the figure represent reference signal resource set #1 and reference signal resource set #2, respectively. Reference signal resource set #1 can correspond to the first resource configuration, and reference signal resource set #2 can correspond to a second resource configuration, i.e., P = 1. The terminal device can periodically send first inference information at the first reporting time to indicate the inference results for reference signal resource set #1 and reference signal resource set #2, respectively.

[0353] As shown in Figure 10(b), the first reporting period differs from the P second reporting periods. Sets #1 and #2 in the figure represent reference signal resource sets #1 and #2, respectively. Reference signal resource set #1 corresponds to the first resource configuration, and reference signal resource set #2 corresponds to the second resource configuration, i.e., P = 1. The first reporting period is an integer multiple of the second reporting period. The terminal device can send first inference information at the first reporting time and second inference information at the second reporting time, distinguished by different line types in the figure.

[0354] As shown in Figure 10(c), the first reporting period differs from the P second reporting periods. Sets #1, #2a, and #2b in the figure represent reference signal resource sets #1, #2a, and #2b, respectively. Reference signal resource set #1 can correspond to a first resource configuration, while reference signal resource sets #2a and #2b correspond to two different second resource configurations. The second reporting periods for these two second resource configurations are different. The figure shows two second reporting periods, period #1 and period #2, corresponding to reference signal resource sets #2a and #2b, respectively, and corresponding to two different second resource configurations: second resource configuration #a and second resource configuration #b, i.e., P = 2. In this context, the second reporting period configured for the second resource configuration #a (e.g., denoted as period #a) is an integer multiple of the second reporting period configured for the second resource configuration #b (e.g., denoted as period #b). The first reporting period is an integer multiple of the second reporting period configured for the second resource configuration #a (i.e., period #a), which is also an integer multiple of the second reporting period configured for the second resource configuration #b (i.e., period #b). The terminal device can send first inference information at the first reporting time and second inference information at the second reporting time. The information indicated by the second inference information reported at different second reporting times is also different, which is distinguished by different line types in the figure. At the second reporting time corresponding to period #a, the second inference information indicates the inference results for reference signal resource set #2a and reference signal resource set #2b. At the second reporting time corresponding to period #2b, the second inference information indicates the inference results for reference signal resource set #2b.

[0355] It should be understood that Figure 10 is for illustrative purposes only and should not impose any limitations on the size of the first reporting period, the P second reporting periods, or the value of P.

[0356] It should also be understood that the first reporting period and the P second reporting periods are both integer multiples of the transmission period of the reference signal resource set configured by the third resource configuration. Therefore, the transmission period of the reference signal resource set configured by the third resource configuration can also be said to be the common divisor of the first reporting period and the P second reporting periods.

[0357] As mentioned earlier, the first and second inference information can be carried through a CSI report, which can be configured by a reporting configuration (such as CSI-ReportConfig). The following details the resource allocation and counting rules for CSI processing under configuration methods one and two. For example, the CSI processing resource is, for instance, a CSI processing unit (CPU).

[0358] In this embodiment, each CSI report corresponds to a period of CSI processing resource occupation. The start time of the CSI processing resource occupation period for each report is the time of the last reference signal received before the CSI reference resource, and the end time can be the reporting time of the CSI report. In other words, the occupation of CSI processing resources can start from the latest RS of the CSI reference resource no later than before the CSI report is reported, and end at the end of the CSI report reporting. If the first reporting period is the same as P second reporting periods, the CSI report only carries the first inference information. Therefore, the end time of the CSI processing resource occupation period is the transmission time of the first inference information. If the first reporting period differs from at least two of the P second reporting periods, or if the first reporting period differs from any one of the P second reporting periods, since the CSI report can carry either first inference information or second inference information, the end time of the corresponding CSI processing resource occupation period is the sending time of the first inference information for a CSI report carrying first inference information, and the end time of the corresponding CSI processing resource occupation period is the sending time of the second inference information for a CSI report carrying second inference information.

[0359] In one possible implementation, the inference of the reference signal resources configured for the first resource configuration and the inference of the reference signal resources configured for the P second resource configurations can be performed by the same model; that is, the first task can be implemented by a single model. Since the P second beam sets corresponding to the P second resource configurations are all subsets of the first beam sets corresponding to the first resource configurations, the model can first infer the reference signal resource sets configured for the first resource configurations to determine the information of each beam in the first beam set. Then, based on the range of the second beam sets corresponding to the reference signal resource sets configured for each second resource configuration, and the inferred information of each beam in the first beam set, the model selects the top beam corresponding to the second beam set.

[0360] Therefore, the model (or the first task) can perform inference on the set of reference signal resources configured in the first resource configuration, without having to perform inference on the set of reference signal resources configured in each of the second resource configurations. Thus, the amount of processing resources occupied by the first CSI report is related to the number of reference signal resources in the set of reference signal resources configured in the first resource configuration, but not to the number of reference signal resources in the set of reference signal resources configured in each of the second resource configurations. In other words, whether the inference is performed on the first resource configuration or on P second resource configurations, the processing resources used for CSI are related to the number of reference signal resources included in the set of reference signal resources configured in the first resource configuration. Optionally, the amount of processing resources occupied by the first CSI report is equal to the number of reference signal resources in the set of reference signal resources configured in the first resource configuration, or the amount of processing resources occupied by the first CSI report is proportional to the number of reference signal resources in the set of reference signal resources configured in the first resource configuration.

[0361] Optionally, prior to step 810, the method further includes step 850: the terminal device sends capability information, which indicates support for determining inference results for a first beam set and multiple fourth beam sets, each of the multiple fourth beam sets being a subset of the first beam set, and the reference signal resource set configured in one of the P second resource configurations corresponding to one of the multiple fourth beam sets. Accordingly, the network device receives the capability information.

[0362] After model training is complete, the terminal device can send capability information to the network device to notify the network device of the capabilities supported by the model. In this embodiment, the capability information can be used to indicate that the model (or the first task) supports determining inference results for a first beam set and multiple fourth beam sets, or supports determining inference results for multiple beam sets with subset relationships.

[0363] The P sets of second beams are subsets of multiple sets of fourth beams. Optionally, the multiple sets of fourth beams are the P sets of second beams. That is, the terminal device sends capability information to the network device. This capability information can be used to indicate that the model (or the first task) supports determining inference results for the first beam set and the P sets of second beams, or supports determining inference results for multiple beam sets with subset relationships, or supports determining inference results for multiple resource configurations that satisfy the first relationship.

[0364] It should be understood that the model may be deployed in a terminal device or in other devices besides the terminal device, and this application does not limit this. It should also be understood that there may be one model or multiple models. If there is one model, the first task can be performed by one model; if there are multiple models, the first task can be performed by multiple models.

[0365] If determining the inference result for a set of reference signal resources is considered as a function, then the above situation of performing the first task through a model can also be regarded as that model having multiple functions.

[0366] Based on the above scheme, network devices can associate multiple resource configurations through a single reported configuration, thereby facilitating terminal devices to determine the first inference information based on the reported configuration. Therefore, network devices do not need to send a separate reported configuration for each resource configuration, reducing configuration overhead.

[0367] Figure 11 is a schematic flowchart of a communication method 1100 provided in another embodiment of this application. For ease of understanding and explanation, the method shown in Figure 11 uses a terminal device as an example of a first device and a network device as an example of a second device, illustrating the implementation flow of the method 1100. Unlike the method 800 shown in Figure 8, method 1100 associates multiple resource configurations through multiple reporting configurations and optimizes the resource occupancy and counting rules for CSI processing.

[0368] As shown in Figure 11, method 1100 includes steps 1110 to 1120. Optionally, it also includes one or more of steps 1130 or 1140. The steps in method 1100 are described in detail below.

[0369] In step 1110, the network device sends multiple reporting configurations, including a first reporting configuration and P second reporting configurations associated with the first reporting configuration. Correspondingly, the terminal device receives these multiple reporting configurations.

[0370] Unlike method 800, each reporting configuration in this embodiment is used to associate one of the first resource configuration or the second resource configuration, as well as the third resource configuration.

[0371] The first reporting configuration can be associated with a first resource configuration and a third resource configuration. The reference signal resource set configured in the first resource configuration corresponds to a first beam set, and the reference signal resource set configured in the third resource configuration corresponds to a third beam set. For a description of the first and third beam sets, please refer to the description in method 800 above, which will not be repeated here. The first reporting configuration can be used to configure an inference report, for example, denoted as the first inference report. This first inference report can be used to report the top beam in the first beam set corresponding to the reference signal resource set configured in the first resource configuration.

[0372] Similar to the first reporting configuration, each of the P second reporting configurations is associated with one of the P second resource configurations and a third resource configuration. The reference signal resources configured in each second resource configuration correspond to a second beam set. For a description of the second beam set, please refer to the description in method 800 above, which will not be repeated here. Each second reporting configuration can also be used to configure an inference report, for example, denoted as a second inference report. The P second reporting configurations can be used to configure P second inference reports. Each second inference report can be used to report the top beam in the second beam set corresponding to the reference signal resource set configured by a second resource configuration.

[0373] Among them, the first resource configuration and P second resource configurations have a first relationship, which can specifically satisfy: the reference signal resource set configured by each of the P second resource configurations is a subset of the reference signal resource set configured by the first resource configuration.

[0374] For example, one possible implementation for associating the multiple reporting configurations is that the first reporting configuration includes the identifiers of P second reporting configurations. Another possible implementation is that each of the P second reporting configurations includes the identifier of the first reporting configuration. Yet another possible implementation is that the first reporting configuration and each of the P second reporting configurations includes the same identifier. Reporting configurations with the same identifier are associated reporting configurations.

[0375] In step 1120, the terminal device determines multiple inference reports based on the multiple reporting configurations.

[0376] The terminal device can determine, based on the multiple reported configurations, the need for inference on the first resource configuration and P second resource configurations, and carry the inference results obtained for each resource configuration through an independent inference report. That is, multiple inference reports corresponding to the multiple reported configurations can be determined. Specifically, determining the inference result for the first resource configuration means determining the inference result for the set of reference signal resources configured in the first resource configuration, which can be inferred based on the measurement results of the reference signal resources configured in the third resource configuration. Similarly, determining the inference result for each second resource configuration means determining the inference result for the set of reference signal resources configured in each second resource configuration, which can be inferred based on the measurement results of the reference signal resources configured in the third resource configuration.

[0377] In one possible implementation, the above inference can be obtained through the same model, which can be trained using method 1400 described below. In other words, the model has the ability to determine inference results for a first resource configuration with a first relation and P second resource configurations, or the model has the ability to determine inference results for reference signal resource sets configured for the first resource configuration and the P second resource configurations respectively. Since the model can obtain inference results by performing an inference task (i.e., the first task described above), it can also be said that the first task has the ability to determine inference results for multiple resource configurations with a first relation (or multiple reference signal resource sets with a subset relation), or the first task has the ability to determine inference results for reference signal resource sets configured for the first resource configuration and the P second resource configurations respectively. Since the model can be deployed on terminal devices or on devices other than terminal devices (such as OTT servers), it can also be said that the device with the model deployed (such as a terminal device or an OTT server) has the ability to support determining the inference result for multiple reference signal resource sets with a first relationship. In other words, the device with the model deployed (such as a terminal device or an OTT server) has the ability to support determining the inference result for reference signal resource sets configured with a first resource configuration and P second resource configurations respectively.

[0378] In this embodiment, the inference results for the first resource configuration and P second resource configurations are similar to the inference results in method 800 above, and can be information about one or more beams, i.e., beam information of the top beam. For a more detailed explanation of the top beam and beam information, please refer to the description in method 800 above, which will not be repeated here.

[0379] Understandably, the model can be deployed on a terminal device, allowing the terminal device to perform model inference and obtain inference results independently. Alternatively, the model can be deployed on other devices besides the terminal device, such as inference nodes. In this case, the terminal device can be called a data collection node, which can collect inference results from the inference node and then determine multiple inference reports based on those results. In other words, the data collection node and the inference node can be the same device or different devices.

[0380] In step 1130, the terminal device sends the multiple inference reports. Correspondingly, the network device receives the multiple inference reports.

[0381] The terminal device can report these multiple inference reports to the network device. For example, the terminal device can send multiple CSI reports to the network device, which can be generated based on multiple CSI-ReportConfigs (i.e., an example of reporting configurations) sent by the network device.

[0382] The terminal device does not necessarily send the first inference information every time it receives it. In the event of insufficient resources, the terminal device may discard the received first inference information or wait until resources are sufficient before sending it; this application does not limit this. In other words, step 1130 is an optional step and does not necessarily have to be performed.

[0383] Optionally, each of the plurality of reporting configurations indicates the reporting period for the configured inference report.

[0384] One possibility is that the multiple reporting configurations indicate the same reporting period. Accordingly, the transmission period of the reference signal resource set configured in the third resource configuration is the same as this reporting period. Alternatively, the reporting period is an integer multiple of the transmission period.

[0385] Optionally, step 1130 specifically includes: the terminal device sending the plurality of inference reports at the first reporting time, the plurality of inference reports including: a first inference report and P second inference reports.

[0386] One possible design for the multiple inference reports is as follows: the first inference report includes information on K beams, which are the top K beams inferred from the reference signal resource set configured for the first resource configuration; the p-th second inference report in the P second inference reports includes K... p Information about the beam, the K p The beam is K, which is inferred from the set of reference signal resources configured for the p-th second resource configuration out of P second resource configurations. p 'One beam, and the Kp The beam is the top K obtained by inference for the set of reference signal resources configured for the p-th second resource configuration. p The beam that is different from the K beams, K, K n and K n ' is a positive integer, and n is an integer from 1 to N.

[0387] In other words, the top beams of the reference signal resource sets configured in the first resource configuration and the P reference signal resource sets configured in the P second resource configurations can be inferred separately. The top beams of different sets are then deduplicated. In this way, the first inference report can include information on K beams, and the p-th second inference report among the P second inference reports can include less than or equal to K beams. p Information about each beam, that is, the aforementioned K p 'Information on each beam, K' p 'Greater than or equal to K' p .

[0388] It should be understood that deduplication means treating overlapping beams as a single beam. Deduplication reduces reporting overhead. In this case, the number of top beams reported for P sets of second beams may not be fixed, and the terminal device can additionally indicate the number of top beams corresponding to the beam information reported for the P sets of second beams.

[0389] It should also be understood that, since the network device knows in advance the subset relationship between each second beam set and the first beam set, it can determine the top beams that overlap between the first beam set and its various subsets (i.e., each second beam set) based on the top K beams indicated by the first inference report. In other words, the network device can determine the deduplicated beams based on the subset relationship between the first beam set and P second beam sets, as well as the first inference report, and then, by combining the P second inference reports, determine the top K beams of the p-th second beam set among the P second beam sets. p Beam.

[0390] K, K1 to K above P One or more of these can be indicated by the network device; for example, K is indicated in the first reporting configuration, K1 to K2. P Indicated in each of the P second reporting configurations; or, K', K1 to K P One or more of these can be predefined by the protocol. In other words, K', K1 to K' above... PThe number of beams indicated by the network device and / or predefined by the protocol can be specified by the network device. For example, the protocol predefines some items, and the network device indicates others; or the protocol predefines all items; or the network device indicates all items. Due to deduplication, the number of beams indicated in the first inference report can be predetermined, but the number of beams indicated in the P second inference reports is not fixed. Therefore, the terminal device can also indicate the number of top beams indicated in each of the P second inference reports, for example, indicating the number of top beams indicated in the p-th inference report in the p-th second inference report.

[0391] Another possible design for these multiple inference reports is: the p-th second inference report in the P second inference reports includes the top K. p The beam information, the first inference report includes information on K' beams, which are the top K' beams identified from the beams in the first beam set excluding the P second beam sets.

[0392] The above K', K1 to K P One or more of these can be indicated by the network device, for example, K' is indicated in the first reporting configuration, K1 to K... P Indicated in each of the P second reporting configurations; or, K', K1 to K P One or more of these can be predefined by the protocol. In other words, K', K1 to K' above... P This can be indicated by the network device and / or predefined by the protocol. For example, the protocol predefines some items, and the network device indicates others; or, the protocol predefines all items; or, the network device indicates all items. Since the number of beams indicated by the first inference report and P second inference reports can be predetermined, that is, the number of beams corresponding to the beam information in the first inference report can be predetermined, and the beam information in the first inference report and the beam information in the P second inference reports do not overlap, more useful beam information can be reported under the same reporting overhead.

[0393] The second possibility is that the multiple reporting configurations are configured with different reporting periods, and the reporting period indicated by the first reporting configuration is an integer multiple of the reporting period indicated by any one of the P second reporting configurations. Correspondingly, the transmission period of the reference signal resource set configured by the third resource configuration can be the same as the minimum period among the multiple second reporting periods configured by the P second reporting configurations, or the reporting period indicated by each of the P second reporting configurations is an integer multiple of the transmission period of the resource configured by the third resource configuration.

[0394] Optionally, step 1130 specifically includes: the terminal device sending a first inference report and P second inference reports at the first reporting time, and sending one or more second inference reports at the second reporting time.

[0395] The first inference report and P second inference reports sent at the first reporting time can be understood by referring to the explanation of multiple inference reports in the first case above, and will not be repeated here. At the second reporting time, since the second reporting periods indicated by each second resource configuration may be the same or different, the second resource configurations corresponding to the second inference reports actually reported in different second reporting times may be different. Therefore, reports can be submitted separately for each second resource configuration. For example, for the reference signal resource set configured in the p-th second resource configuration, the top K inference results are reported. p Information about the beam. Regarding K1 to K... P The value of can be determined by referring to the first design, which can be indicated by the network device or predefined by the protocol, and will not be elaborated further.

[0396] Furthermore, the first inference report and the P second inference reports can also be associated. For example, the first inference report contains the identifiers of the P second inference reports, or each of the P second inference reports contains the identifier of the first inference report. By associating the first inference report and the P second inference reports, the network device can understand that the first inference report and the P second inference reports were obtained by performing inference tasks based on the same model.

[0397] For example, each of the first reporting configuration and the P second reporting configurations described above can be a CSI reporting configuration (CSI-ReportConfig), or it can also be called a CSI report configuration. Each of the first inference report and the P second inference reports can be a CSI report.

[0398] In this embodiment, since the multiple reporting configurations (including a first reporting configuration and P second reporting configurations) are associated, the model (or the first task) can perform inference for the reference signal resource set configured by the first resource configuration, without having to perform inference for the reference signal resource set configured by each of the second resource configurations. Therefore, the processing resources occupied by the multiple reporting configurations are related to the number of reference signal resources contained in the reference signal resource set configured by the first resource configuration, and are independent of the number of reference signal resources in the reference signal resource sets configured by the second resource configurations. In other words, whether the inference is performed for the first resource configuration or for the P second resource configurations, the CSI processing resources occupied are related to the number of reference signal resources contained in the reference signal resource set configured by the first resource configuration. Optionally, the amount of processing resources occupied by the multiple reporting configurations is equal to the number of reference signal resources in the reference signal resource set configured by the first resource configuration, or the amount of processing resources occupied by the multiple reporting configurations is proportional to the number of reference signal resources in the reference signal resource set configured by the first resource configuration.

[0399] In one possible design, for each of the multiple reporting configurations, only one set of CSI processing resources is needed. That is, the multiple reporting configurations share a common set of CSI processing resources, or in other words, this common set of CSI processing resources is used to process the CSI reports corresponding to the multiple reporting configurations. Each CSI report corresponds to a period of CSI processing resource usage. The start time of this period is the time of the last reference signal received before the CSI reference resource, and the end time can be the reporting time of the CSI report. In other words, the usage of CSI processing resources can start from the latest RS (Reference Signal) of the CSI reference resource no later than the time of the CSI report reporting and end with the reporting of the CSI report.

[0400] If the first reporting period and P second reporting periods are all the same, the terminal device can simultaneously report the first inference report and P second inference reports. Therefore, the end time of the period when the common CSI processing resources are occupied is the sending time of the first inference report and the P second inference reports. This can be understood as the CSI processing resources occupied by these multiple reporting configurations being the same as the CSI processing resources occupied when configuring any one of these multiple reporting configurations individually. If the first reporting period differs from at least two of the P second reporting periods, since the CSI report can carry both first and second inference information, the terminal device can simultaneously report the first inference report and the P second inference reports, or report one or more second inference reports individually. For the case of simultaneously reporting the first inference report and the P second inference reports, the end time of the CSI processing resource occupation period can be the sending time of the first inference report, i.e., the first reporting time. For the case of reporting second inference reports but not the first inference report, the end time of the CSI processing resource occupation period can be the sending time of the second inference report, i.e., the second reporting time. This can be understood as the CSI processing resources occupied by these multiple reporting configurations being the same as those occupied by configuring the reporting configuration with the shortest period among these multiple reporting configurations individually.

[0401] Figure 12 is a schematic diagram illustrating the time period of processing resource occupation when the first reporting cycle and the second reporting cycle are the same, as shown in an embodiment of this application.

[0402] Figure 12, using P=1 as an example, illustrates the case where the first reporting cycle and the second reporting cycle are the same. That is, the network device configures the same reporting cycle for both the first and second resource configurations, namely the first reporting cycle. The time period for resource occupancy in the figure can be defined as starting from the time of the last reference signal received before the CSI reference resource and ending at the reporting time of the first and second inference reports.

[0403] Figure 13 is a schematic diagram illustrating the time period of processing resource occupation when the first reporting cycle and the second reporting cycle are different, as shown in an embodiment of this application.

[0404] Figure 13, using P=1 as an example, illustrates the case where the first reporting period and the second reporting period are different. That is, the network device configures different reporting periods for the first resource configuration and the second resource configuration, namely the first reporting period and the second reporting period. The specific time periods for occupying processing resources in the figure are as follows: If the first inference report and the second inference report are sent together, the time period for occupying processing resources starts from the time of the last reference signal received before the CSI reference resource and ends at the reporting time of the first and second inference reports; if the second inference report is sent alone, the time period for occupying processing resources starts from the time of the last reference signal received before the CSI reference resource and ends at the reporting time of the second inference report.

[0405] Optionally, prior to step 1110, the method further includes step 1140: the terminal device sends capability information, which indicates support for determining inference results for the first beam set and multiple fourth beam sets. Accordingly, the network device receives the capability information.

[0406] For more detailed information on step 1140, please refer to the relevant description of step 850 in method 800, which will not be repeated here.

[0407] Based on the above scheme, the network device associates multiple resource configurations through multiple reported configurations, thereby associating a first beam set with a subset relationship with P second beam sets. This allows the execution of the first task to consider this subset relationship, thus obtaining the inference result of the reference signal resource set configured for the multiple resource configurations. In this embodiment, because the resource counting rules are optimized, the resource counting of the third resource set can be saved. The optimization of the resource occupancy and counting rules also saves processing resources.

[0408] In methods 800 and 1100 described above, the first task can be performed by the same model, which can be trained. The data collection process for model training will be explained in detail below with reference to Figure 14.

[0409] To improve the performance of predictions for subsets of the first beam set and reduce training overhead by leveraging subset relationships, this application also provides a method for data collection prior to model training. It is understood that this method can also be implemented in conjunction with methods 800 and 1100.

[0410] Figure 14 is a schematic flowchart of a communication method 1400 provided in another embodiment of this application. For ease of understanding and explanation, the method shown in Figure 14 uses a terminal device as an example of a third device and a network device as an example of a fourth device, illustrating the implementation flow of method 1400. It should be understood that the third device and the first device in methods 800 and 1100 above can be the same device or different devices; the fourth device and the second device in methods 800 and 1100 above can be the same device or different devices. In other words, the terminal device in this embodiment and the terminal device in methods 800 and 1100 above can be the same terminal device or different terminal devices; the network device in this embodiment and the network device in methods 800 and 1100 above can be the same network device or different network devices.

[0411] As shown in Figure 14, method 1400 includes steps 1410 and 1420. Optionally, it also includes one or more of steps 1430 or 1440. The steps in method 1400 are described in detail below.

[0412] In step 1410, the network device sends first information indicating a first beam set. Correspondingly, the terminal device receives this first information.

[0413] The first beam set can be considered as a full beam set, such as Set A. It should be understood that the first beam set can be the same beam set as the first beam set in method 800 and method 1100. Please refer to the relevant explanation of the first beam set above, which will not be repeated here.

[0414] Optionally, the first beam set corresponds to the first reference signal resource set. One possible way the first information indicates the first beam set is that it configures the first reference signal resource set, which includes X reference signal resources, and the first beam set includes X beams, with each of the X reference signal resources corresponding one-to-one with the X beams, where X is a positive integer. For example, the first information includes identifiers for the X reference signal resources, and these identifiers correspond one-to-one with the X reference signal resources included in the first reference signal resource set.

[0415] As explained in the terminology section above, a beam can be considered a spatial resource for reference signals. Therefore, reference signal resources used to transmit reference signals can have a one-to-one correspondence with beams. In this embodiment, the X reference signal resources included in the first reference signal resource set can correspond one-to-one with the X beams in the first beam set. The network device can configure the first reference signal resource set through the first information, thereby configuring the first beam set.

[0416] For example, the first information could be a CSI-ResourceConfig.

[0417] In addition to the first beam set, the network device can also configure a third beam set. The method further includes: the network device sending seventh information to indicate the third beam set. This third beam set is used to obtain the input data corresponding to the training. The third beam set can be considered as a set of partial beams from the full beam set, or a beam set different from the full beam set, such as Set B. It should be understood that this third beam set can be the same beam set as the third beam set in methods 800 and 1100; see the above explanation regarding the third beam set for further details.

[0418] Optionally, the third beam set corresponds to the third reference signal resource set. The method further includes: the network device sending seventh information, which is used to configure the third reference signal resource set, the third reference signal resource set including Z reference signal resources, and the third beam set including Z beams, wherein the Z reference signal resources correspond one-to-one with the Z beams, and Z is a positive integer. For example, the seventh information includes identifiers of the Z reference signal resources, and the identifiers of the Z reference signal resources correspond one-to-one with the Z reference signal resources included in the third reference signal resource set. For example, the seventh information can be another CSI-ResourceConfig. Further, the first information and the seventh information can be associated through a reporting configuration. For example, the reporting configuration can be CSI-ReportConfig. Through association, the terminal device can determine that the first beam set corresponding to the first reference signal resource set configured by the first information and the third beam set corresponding to the third reference signal resource set configured by the seventh information are used for training the first task.

[0419] As an example, a network device can configure a CSI report for training data collection, for instance, via a CSI-ReportConfig. In the CSI report for training data collection, the network device can configure measurement resources, including resources for Set A (i.e., an example of a first set of reference signal resources) and resources for Set B (i.e., an example of a third set of reference signal resources). The terminal device can perform measurements based on the configured measurement resources, and the measured data can be used for training. For example, a CSI-ResourceConfig can be configured for each of the resources in Set A and Set B. A CSI-ResourceConfig can be used to configure a set of reference signal resources, each set of reference signal resources including one or more associated measurement resources (i.e., reference signal resources, such as NZP-CSI-RS resources or SSB resources).

[0420] Optionally, the first reference signal resource set and the third reference signal resource set are associated through a reporting configuration. Alternatively, the first information used to configure the first reference signal resource set and the seventh information used to configure the third reference signal resource set are associated through a reporting configuration.

[0421] For example, a network device can configure a CSI-ReportConfig for training data collection (i.e., an example of a reporting configuration), in which the CSI-ReportConfig simultaneously associates the CSI-ResourceConfig corresponding to the resources of Set A and the resources of Set B.

[0422] Optionally, the aforementioned first reference signal resource set and third reference signal resource set are associated through two reporting configurations. Alternatively, the first information used to configure the first reference signal resource set and the seventh information used to configure the third reference signal resource set are associated through two reporting configurations.

[0423] For example, a network device can configure two CSI-ReportConfigs for training data collection (i.e., an example of a reporting configuration). One CSI-ReportConfig is associated with the CSI-ResourceConfig corresponding to resources in Set A, and the other CSI-ReportConfig is associated with the CSI-ResourceConfig corresponding to resources in Set B. These two CSI-ReportConfigs also need to be linked. For example, the identifier of the CSI-ReportConfig associated with Set A can be indicated in the CSI-ReportConfig associated with Set B, or the identifier of the CSI-ReportConfig associated with Set B can be indicated in the CSI-ReportConfig associated with Set A.

[0424] Optionally, the reported configuration may also include one or more associated IDs.

[0425] Association identifiers can be used to indicate the beam properties of reference signal resources. Reference signal resource sets with the same association identifier have the same or similar beam properties. Beam properties may include, but are not limited to, beam directivity, energy concentration, and coherence, and can be affected by transmit power, elevation angle, azimuth, beam pointing, etc. Therefore, when the third beam set is a subset of the first beam set, the reported configuration includes one association identifier; when the third beam set is not a subset of the first beam set, the reported configuration includes two association identifiers: one assigned to the first reference signal resource set and the other assigned to the third reference signal resource set.

[0426] In step 1420, the network device sends Q pieces of second information, which are used to indicate Q sets of fourth beams, each of the Q pieces of second information indicating one of the Q sets of fourth beams. Correspondingly, the terminal device receives the Q pieces of second information.

[0427] Each of the Q sets of fourth beams is a subset of the first set of beams. Therefore, the first set of beams and the Q sets of fourth beams can be used for training the first task, or in other words, for model training, so that the trained model can be used to perform the first task.

[0428] It should be understood that the P second beam sets in methods 800 and 1100 above can be subsets of the Q fourth beam sets; in other words, each second beam set can be the same as a fourth beam set. Alternatively, P = Q, meaning the P second beam sets and the Q fourth beam sets are the same P beam sets. In fact, both the second and fourth beam sets are subsets of the first beam set. In this paper, the terms "second beam set" and "fourth beam set" are used for ease of distinguishing between the beam subsets used for training and those used in the inference process. For an understanding of the fourth beam set, please refer to the relevant explanation of the second beam set above; it will not be repeated here.

[0429] Optionally, the Q pieces of second information are used to configure Q sets of fourth reference signal resources, wherein the Q sets of fourth reference signal resources correspond one-to-one with the Q sets of fourth beams.

[0430] The relationship between the Q sets of fourth reference signal resources and the Q sets of fourth beams can be understood by referring to the relationship between the reference signal resource sets and the second beam sets configured in the second resource configuration above, and will not be repeated here. In this embodiment, the network device configures the Q sets of fourth reference signal resources through Q pieces of second information, thereby realizing the configuration of the Q sets of fourth beams.

[0431] As mentioned earlier, in one possible design, the first information (e.g., a CSI-ResourceConfig) used to configure the first reference signal resource set and the seventh information (e.g., another CSI-ResourceConfig) used to configure the third reference signal resource set can be associated through a reporting configuration (e.g., CSI-ReportConfig). In this case, the Q second information used to configure the Q fourth reference signal resource sets can also be associated through the same CSI-ReportConfig.

[0432] In another possible design, the first information (such as a CSI-ResourceConfig) used to configure the first reference signal resource set and the seventh information (such as another CSI-ResourceConfig) used to configure the third reference signal resource set can be associated with two (such as CSI-ReportConfig). In this case, the Q second information used to configure the Q fourth reference signal resource sets can be associated with the first information through the same reporting configuration (such as CSI-ReportConfig), or they can be associated through different reporting configurations, without limitation.

[0433] For example, the Q pieces of second information are Q CSI-ResourceConfig, each corresponding to one of the Q sets of fourth reference signal resources.

[0434] The following will detail the specific implementation of the Q second pieces of information used to indicate the Q second beam sets.

[0435] In one possible implementation, the Q second pieces of information can indicate the Q second beam sets by indicating the Q fourth reference signal resource sets.

[0436] Optionally, each of the Q fourth reference signal resource sets is a subset of the first reference signal resource set. Optionally, the q-th fourth reference signal resource set among the Q fourth reference signal resource sets includes Y. q A reference signal resource, the Y q Y in X reference signal resources and X reference signal resources q The reference signal resources are identical, meaning they correspond to the same beam. For example, each of the Q pieces of second information can be used to indicate the identifier of the reference signal resources included in a fourth reference signal resource set, and the identifier of the reference signal resources included in each fourth reference signal resource set is contained in the identifier of the reference signal resources included in the first reference signal resource set.

[0437] For example, a network device can be configured with Q CSI-ResourceConfigs, and the reference signal resource set associated with each CSI-ResourceConfig is a subset of the first reference signal resource set. For instance, the NZP-CSI-RS-ResourceId or SSB-Index associated with one or more CSI-ResourceConfigs (i.e., two instances of the identifier of the reference signal resource) are all in the NZP-CSI-RS-ResourceId or SSB-Index associated with the CSI-ResourceConfig corresponding to the first reference signal resource set.

[0438] Optionally, the q-th fourth reference signal resource set in the Q fourth reference signal resource sets includes Y. q A reference signal resource, the Y q Y in X reference signal resources and X reference signal resources qEach reference signal resource has a one-to-one mapping relationship, and two reference signal resources with this mapping relationship correspond to the same beam. In other words, each of the Q sets of fourth reference signal resources is not necessarily a subset of the first set of reference signal resources; that is, the reference signal resources in the fourth set of reference signal resources may not be the same as those in the first set of reference signal resources, but they correspond to the same beam. For example, a beam might correspond to a reference signal resource in the first set with an identifier of 0, while the same beam might correspond to a reference signal resource in the fourth set with an identifier of 64. In this case, the second information can explicitly or implicitly indicate the mapping relationship between two reference signal resources corresponding to the same beam.

[0439] For example, a network device can configure Q CSI-ResourceConfigs and indicate the identifier of the reference signal resource associated with each CSI-ResourceConfig. For instance, the NZP-CSI-RS-ResourceId or SSB-Index associated with these Q CSI-ResourceConfigs (i.e., two examples of the reference signal resource identifiers) may differ from the NZP-CSI-RS-ResourceId / SSB-Index associated with the CSI-ResourceConfig corresponding to Set A. However, the resources associated with these Q CSI-ResourceConfigs have a one-to-one mapping relationship with a portion of the resources in the first set of reference signal resources. This mapping relationship can be indicated by the network device, for example, in the Q pieces of second information, or it can be determined according to preset rules.

[0440] For ease of understanding and explanation, the following text does not use a second piece of information as a general example to describe the determination of this mapping relationship.

[0441] As an example, the network device can indicate the identifiers of reference signal resources with mapping relationships in the second information. For instance, it can add an information element (IE) to CSI-ResourceConfig to indicate the identifiers of reference signal resources in the first reference signal resource set that have mapping relationships with each reference signal resource in the fourth reference signal resource set configured in the second information. For example, for reference signal resource b in the fourth reference signal resource set configured in the second information, it can indicate that resource a in the first reference signal resource set has a mapping relationship with reference signal resource b.

[0442] As an example, the preset rule could be: the resource identified as 'a' in the first reference signal resource set has a mapping relationship with the resource identified as 'a+b' in the fourth reference signal resource set; that is, there is a mapping relationship between the two reference signal resources corresponding to the two identifiers with a difference of 'b'. For example, if 'a' is 0 and 'b' is 64, then the resource identified as '0' in the first reference signal resource set has a mapping relationship with the resource identified as '64' in the fourth reference signal resource set.

[0443] As an example, the preset rule could also be: the identifiers of multiple reference signal resources in the first reference signal resource set and the multiple reference signal resources in the fourth reference signal resource set are mapped one-to-one according to a predefined order. Here, resources with identifiers taking predefined values ​​(such as "None" or "NULL") represent resources without actual reference signals. The predefined order could be, for example, the order of the reference signal resources in the first and fourth reference signal resource sets from front to back, or from back to front, etc. For instance, the first reference signal resource set includes 64 reference signal resources, identified as 0 to 63; the fourth reference signal resource set includes 64 reference signal resources, of which 32 resources are identified as 64 to 95, and 32 resources are identified as "NULL". Based on the order of the reference signal resources in the first and fourth reference signal resource sets, the following mapping relationship can be obtained: Resources marked as 0 in the first reference signal resource set are mapped to resources corresponding to the same identifier in the first reference signal resource set for each non-NULL identifier in the fourth reference signal resource set. For example, if the first identifier in the fourth reference signal resource set is NULL, the second is 64, and the third is 1, then the resources marked 64 and 1 are mapped. This allows us to determine which resource in the first reference signal resource set corresponds to each non-NULL identifier in the fourth reference signal resource set. The 32 NULL values ​​in the fourth reference signal resource set represent positions with no actual resources; therefore, they are not mapped to any resource in the first reference signal resource set.

[0444] It should be understood that the preset rules exemplified above are merely examples and should not constitute any limitation on this application.

[0445] By determining the mapping relationship between resources in the first reference signal resource set and resources in the fourth reference signal resource set, the terminal device can determine which resources correspond to the same beam.

[0446] In another possible implementation, the Q second pieces of information can directly indicate the Q fourth beam sets.

[0447] For example, each of the Q second pieces of information can be a bitmap, which includes X indicator bits. These X indicator bits correspond one-to-one with the X beams in the first beam set. Each of the X indicator bits in the q-th second piece of information in the Q second pieces of information is used to indicate whether a beam in the corresponding first beam set belongs to the fourth beam set indicated by the q-th second piece of information. In other words, the beams included in each of the Q second beam sets are indicated by a bitmap of length X bits. For example, an indicator bit of "0" indicates that the beam in the first beam set corresponding to that indicator bit does not belong to the fourth beam set, and an indicator bit of "1" indicates that the beam in the first beam set corresponding to that indicator bit belongs to the fourth beam set. For example, for a bitmap, if the first indicator bit is "0", it means that the first beam in the first beam set does not belong to the fourth beam set indicated by the bitmap. If the second indicator bit is "1", it means that the second beam in the first beam set does not belong to the fourth beam set indicated by the bitmap.

[0448] It should be understood that the methods listed above for indicating the Q sets of fourth beams are merely examples and should not constitute any limitation on this application.

[0449] It should also be understood that since the first beam set corresponds to the first reference signal resource set, and the Q fourth beam sets correspond to the Q fourth reference signal resource sets, the prediction of the beam set (including one or more beam sets among the first beam set and the Q fourth beam sets) in the following text can also be understood as the prediction of the corresponding reference signal resource set (including one or more reference signal resource sets among the first reference signal resource set and the Q fourth reference signal resource sets).

[0450] Optionally, the method further includes: the network device sending third information, the third information indicating that the beam set corresponding to the output result of the first task includes a first beam set and Q fourth beam sets. Accordingly, the terminal device receives the third information.

[0451] The network device can use this third information to instruct the output of prediction results for the first beam set and the Q fourth beam sets. As mentioned earlier, the first beam set and the Q fourth beam sets are used for training the first task. Therefore, based on this third information, the terminal device can determine that the beam set corresponding to the output result of the first task includes the first beam set and the Q fourth beam sets.

[0452] As described in methods 800 and 1100 above, the inference result of the model performing the first task includes the information of the top beams inferred from the first beam set and P second beam sets. Therefore, during training, the output of the first task is also the information of the top beams inferred from the first beam set and Q fourth beam sets. During training, the actual measured values ​​of the model's output of the first task can be used as the corresponding label data for training.

[0453] It should be understood that the beam set corresponding to the output result of the first task, including the first beam set and Q second beam sets, can also be predefined by the protocol. In this case, the network device does not need to send the third information. In other words, step 1430 is an optional step.

[0454] For example, prediction is required by default for the first reference signal resource set and the Q fourth reference signal resource sets. Alternatively, the protocol specifies that prediction is required for all resource sets except Set B. Or, prediction is required for all resource sets related to Set A.

[0455] The third information could, for example, include indication information (e.g., true / false) for the first beam set and the Q fourth beam sets, indicating whether prediction is needed for that beam set. Alternatively, the third information could be a (Q+1)-bit bitmap, corresponding one-to-one with the first beam set and the Q fourth beam sets, indicating whether prediction is needed for the corresponding beam set through the value of each indication bit. Alternatively, the third information could include identifiers of reference signal resource sets that need prediction, indicating which reference signal resource sets need prediction. Alternatively, the third information could include indication information (e.g., true / false) to indicate whether prediction is needed for all resource sets (including the first beam set and the Q fourth beam sets). Alternatively, the protocol could predefine that prediction is needed for the first beam set, and the third information could indicate which of the Q fourth beam sets need prediction.

[0456] Optionally, the method further includes: the network device sending fourth information, the fourth information being used to indicate one or more beam set groups, any one of the one or more beam set groups including a first beam set and at least one of Q fourth beam sets, or including at least two of Q fourth beam sets.

[0457] In other words, a beam set can be a combination of some or all of the first beam set and the Q fourth beam sets. Using the fourth information, the network device can indicate which beam sets can be combined.

[0458] For example, the fourth information can indicate the identifiers of one or more sets of reference signal resources. The reference signal resource sets corresponding to the identifiers of the same set of reference signal resources can be regarded as a group of reference signal resource sets, and the corresponding beam sets can also be regarded as a group of beam sets.

[0459] Optionally, the method further includes: the network device sending fifth information, which is also used to indicate the performance requirements of the output results of the first task. Accordingly, the terminal device receives the fifth information.

[0460] As mentioned earlier, the beam set corresponding to the output results of the first task includes a first beam set and Q fourth beam sets. The performance requirements for the output results of the first beam set and the performance requirements for the output results of the Q fourth beam sets can be the same or different.

[0461] The network device can specify performance requirements for the output results for the first beam set and the Q fourth beam sets, i.e., specify (Q+1) performance requirements, which can be the same or different. Alternatively, the network device can specify a single performance requirement for the first beam set and the Q fourth beam sets, meaning the performance requirement for the output results of the first beam set is the same as the performance requirement for the output results of the Q fourth beam sets. Performance requirements may include, for example, requirements for prediction accuracy or precision, or requirements for latency, etc., without limitation.

[0462] Optionally, the association identifiers of the Q fourth beam sets may be the same as or different from those of the first beam set.

[0463] In other words, the Q sets of fourth beams can have the same beam characteristics as the first beam set, or they can have different beam characteristics.

[0464] Optionally, the method further includes: the network device sending a sixth message, the sixth message including Q associated IDs, the Q associated IDs corresponding one-to-one with Q fourth beam sets; or, the sixth message including one associated ID.

[0465] For an explanation of this association identifier, please refer to the description in step 1410 above, which will not be repeated here.

[0466] If the Q fourth beam sets have the same beam characteristics as the first beam set, the sixth information can be used to indicate one association identifier, or it can indicate Q association identifiers that are identical and the same as the association identifier of the first beam set. If some or all of the Q fourth beam sets have different beam characteristics than the first beam set, the sixth information can be used to indicate Q association identifiers that correspond to the Q fourth beam sets. If the association identifier of a fourth beam set is the same as the association identifier of the first beam set, it indicates that the fourth beam set and the first beam set have the same beam characteristics; if the association identifier of a fourth beam set is different from the association identifier of the first beam set, it indicates that the fourth beam set and the first beam set have different beam characteristics.

[0467] In practical applications, network devices may adjust beam direction, elevation angle, etc., according to requirements, causing changes in beam characteristics. Therefore, the first beam set configured by the network device at different times may have different beam characteristics. In this case, some subsets of the first beam set may also be adjusted accordingly. For example, if a beam in the first beam set changes, one subset of the first beam set (e.g., denoted as subset 1, i.e., one of the Q fourth beam sets) remains unaffected, while another subset (e.g., denoted as subset 2, the other of the Q fourth beam sets) changes accordingly. Therefore, the network device can assign corresponding association identifiers to the Q fourth beam sets to indicate beam sets with the same beam characteristics. For example, in the above example, the network device can assign new association identifiers to the first beam set and subset 1, without needing to assign a new association identifier to subset 1. In this way, the terminal device can determine the subset relationships between beam sets.

[0468] Among the information items above, the second, third, fourth, fifth, and sixth information items are all related to the Q sets of fourth beams. These Q sets of fourth beams are subsets of the first beam set, and therefore can be called training auxiliary information for the subset.

[0469] Optionally, the method further includes step 1430: the terminal device performs training for the first task based on the first information, the second information, and the third measurement result.

[0470] The third measurement result is obtained based on measurements of a third set of reference signal resources, which includes the input data corresponding to the training of the first task.

[0471] The terminal device responds to the configuration of the first and second information, and based on its own capabilities, it can have the following various training results.

[0472] Training Result 1: For each set of reference signal resources that needs to output prediction results, a unified model is trained.

[0473] Training Result 2: For each set of reference signal resources that needs to output prediction results, train multiple different models.

[0474] The following explanations will elaborate on these two training results respectively.

[0475] Training Result 1: For each set of reference signal resources that needs to output prediction results, a unified model is trained.

[0476] For example, step 1430 may specifically include:

[0477] The terminal device measures the first set of reference signal resources and obtains a first measurement result;

[0478] The terminal device measures the third reference signal resource set and obtains the third measurement result;

[0479] The terminal device performs training for the first task based on the first measurement result and the third measurement result; wherein the third measurement result includes the input data corresponding to the training of the first task, and the first measurement result includes the label data corresponding to the training of the first task.

[0480] The third measurement result may include measurement values ​​obtained based on the third reference signal resource set. The third measurement result includes the input data corresponding to the training of the first task; that is, some or all of the measurement values ​​in the third measurement result can be used as input data. The first measurement result may include measurement values ​​obtained based on the first reference signal resource set. The first measurement result includes the label data corresponding to the training of the first task; that is, some or all of the measurement values ​​in the first measurement result can be used as label data. It can be understood that since the Q fourth reference signal resource sets are subsets of the first reference signal resource set, the measurement value corresponding to each resource in the Q fourth reference signal resource sets can also be obtained based on the first measurement result. Therefore, the terminal device can measure only the first reference signal resource set and use some or all of the measurement values ​​in the first measurement result as label data.

[0481] In another implementation, the terminal device can perform actual measurements on one or more of the Q sets of fourth reference signal resources to obtain more accurate measurement values ​​as label data. Optionally, step 1430 further includes: the terminal device measuring one or more of the Q sets of fourth reference signal resources to obtain one or more second measurement results; correspondingly, the terminal device performs training for the first task based on the first measurement results and the third measurement results, including: the terminal device performing training for the first task based on the first measurement results, one or more second measurement results, and the third measurement results; wherein the one or more second measurement results include the label data corresponding to the training of the first task. That is, both the first measurement results and one or more second measurement results are used as label data for training.

[0482] The q-th measurement result among the Q second measurement results may include a measurement value obtained based on the q-th fourth reference signal resource set among the Q fourth reference signal resource sets. One or more second measurement results include label data corresponding to the training of the first task; that is, some or all of the measurement values ​​from the one or more second measurement results are used as label data.

[0483] In other words, the terminal device can use the third measurement result as input data for the model and the first measurement result as label data. Optionally, it can also use one or more second measurement results as label data to train the model. Since the aforementioned training auxiliary information for subsets has been considered during the training process, the performance of subsequent inference can be better guaranteed.

[0484] As an example, the input data and output results of the model during the training process of the first task are shown in Table 3.

[0485] Table 3

[0486] The input data can be a third measurement result. The output results can include prediction results for a first set of reference signal resources and prediction results for Q sets of fourth reference signal resources. The prediction result for each set of reference signal resources can be a prediction value obtained by the model based on the third measurement result for that set of reference signal resources. The label data can be some or all of the following measurement results: the first measurement result and Q sets of second measurement results. For example, one possible implementation of step 1430 is as follows: The terminal device uses some or all of the measurement values ​​in the third measurement result as input data for the initial model, and performs inference for the first set of reference signal resources and one or more sets of the Q sets of fourth reference signal resources respectively, to obtain prediction results corresponding to one or more sets of second measurement results. The terminal device calculates the difference between the predicted and measured values ​​for the reference signal resources in the first reference signal resource set and the Q fourth reference signal resource sets (i.e., the difference between the inference result of the initial AI model and the corresponding label data). Specifically, it calculates the value of the loss function and updates the parameters in the initial AI model based on the loss function value, minimizing the difference between the inference result obtained by the updated AI model and the corresponding label data, i.e., minimizing the loss function. Repeating this process yields an AI model that meets the target requirements.

[0487] Training Result 2: For each set of reference signal resources that needs to output prediction results, train multiple different models.

[0488] For example, for the configured first beam set, the terminal device can train one model; for the configured Q fourth beam sets, the terminal device can train one model for each, i.e., train Q models. Alternatively, it can use training assistance information to merge two or more similar fourth beam sets (e.g., two or more fourth beam sets with overlapping beams) and train one model, i.e., train fewer than Q models. Optionally, the terminal device can first train a model for the first beam set, and then fine-tune the parameters based on this model to obtain one or more models for the Q fourth beam sets.

[0489] As an example, the input data and output results of the model during the training process of the first task are shown in Tables 4 and 5.

[0490] The terminal device can train one model for the configured first set of reference signal resources; and train one or more models for the configured Q sets of fourth reference signal resources. The model training process is similar to the process of training one model in Training Result 1 above, and will not be described again.

[0491] As an example, Table 4 shows an example of training two models. In the training of model #1, the input data can be the third measurement result, the output can be the prediction result for the first set of reference signal resources, and the label data can be some or all of the measurement values ​​from the first measurement result. In the training of model #2, the input data can be the third measurement result, the output can be the prediction result for Q sets of fourth reference signal resources, and the label data can be some or all of the following measurement results: the first measurement result and Q second measurement results.

[0492] Table 4

[0493] As an example, Table 5 shows an example of training multiple models. During the training of model #1, the input data can be the third measurement result, the output result can be the prediction result for the first reference signal resource set, and the label data can be some or all of the measurement values ​​from the first measurement result. During the training of model #q from model #1 to model #Q, the input data can be the third measurement result, the output result can be the prediction result for the q-th reference signal resource set out of Q fourth reference signal resource sets, and the label data can be some or all of the measurement values ​​from the q-th second measurement result out of Q second measurement results.

[0494] Table 5

[0495] As mentioned earlier, the model can be deployed on a terminal device, so the training of the first task can be performed by the terminal device; the model can also be deployed on a device other than the terminal device, so the training of the first task can also be performed by other devices (e.g., referred to as training nodes), in which case the terminal device can be regarded as a data collection node. In other words, the training node and the data collection node can be the same device or different devices. Therefore, for the terminal device, step 1430 is an optional step. However, it is understood that if the training of the first task is performed by other devices, it can also be completed by referring to the method provided in step 1430 above.

[0496] Optionally, the method further includes step 1440: the terminal device sends capability information, which is used to indicate that the first task supports determining inference results for the first beam set and Q fourth beam sets. Accordingly, the network device receives the capability information.

[0497] Since each of the Q fourth beam sets is a subset of the first beam set, this capability information can also be said to be used to instruct the first mission to support determining inference results for multiple sets of reference signal resources with a first relationship (or subset relationship).

[0498] This capability information can be understood by referring to the capability information in method 800 above, and will not be repeated here. The difference is that, in this embodiment, the first task supports determining the inference result for the first beam set and Q fourth beam sets, which can be implemented by one model or by multiple models, and this application does not limit this.

[0499] Based on the above scheme, by adding training auxiliary information for subsets during the training data collection process, the training of the first task considers not only the first beam set but also its subsets (i.e., the aforementioned Q fourth beam sets), thereby improving the model's prediction performance for subsets. Furthermore, by utilizing the relationship between the first beam set and its subsets, joint training can be performed for multiple beam sets requiring prediction. For example, only one model can be trained for the first beam set and its subsets, or only one model can be trained for some closely related subsets. Alternatively, a model can be trained first for the first beam set, and then retrained or fine-tuned for its subsets, eliminating the need to train the model from scratch and reducing training overhead.

[0500] It should be understood that methods 800 and 1100 provided above can be used in conjunction with method 1400. For example, method 1400 can be executed before method 800 to obtain a model that supports determining inference results for multiple sets of reference signal resources with a first relationship, and then the model can be used for inference in method 800 to obtain first inference information and second inference information.

[0501] It should also be understood that the above description of various embodiments uses the example of each resource configuration being able to configure one set of reference signal resources. In practical applications, each resource configuration is not limited to configuring one set of reference signal resources, but can also configure multiple sets of reference signal resources; that is, each resource configuration can be used to configure one or more sets of reference signal resources. In the case where one resource configuration is used to configure multiple sets of reference signal resources, the methods 800, 1100, and 1400 provided above can also be applied. For the sake of brevity, further examples are not provided here.

[0502] The methods provided in the embodiments of this application have been described in detail above with reference to several accompanying drawings. The apparatus provided in the embodiments of this application will now be described with reference to the accompanying drawings.

[0503] Figures 15 to 17 are schematic block diagrams of possible apparatuses provided in embodiments of this application. These apparatuses can be used to implement the functions of the first or second apparatus in the above method embodiments, and thus can also achieve the beneficial effects of the above method embodiments.

[0504] Figures 15 and 16 are schematic diagrams of possible communication devices provided in the embodiments of this application. These communication devices can be used to implement the functions of the terminal devices and network devices in the above method embodiments, and thus can also achieve the beneficial effects of the above method embodiments. In the embodiments of this application, the communication device can be the access network device as shown in Figure 1 or Figure 2, the core network device as shown in Figure 3, the access network node as shown in Figure 4, or the terminal device as shown in Figure 1 or Figure 2, or the terminal as shown in Figure 3 or Figure 4.

[0505] As shown in Figure 15, the communication device 1500 includes a processing module 1510 and a communication module 1520. The communication device 1500 is used to implement the functions of the terminal device in the method embodiments shown in Figures 8, 11, or 14.

[0506] In an optional embodiment, the communication device 1500 is used to implement the functions of the terminal device or network device of the method embodiment shown in FIG8.

[0507] When the communication device 1500 is used to implement the functions of the terminal device in the method embodiment shown in FIG8: the communication module 1520 is used to receive a reported configuration, which is associated with multiple resource configurations, including: a first resource configuration, P second resource configurations and a third resource configuration. Each of the multiple resource configurations is used to configure a set of reference signal resources. The P second resource configurations and the first resource configuration satisfy the following: the beam set corresponding to the reference signal resource set configured by each second resource configuration is a subset of the beam set corresponding to the reference signal resource set configured by the first resource configuration, where P is a positive integer; the processing module 1510 is used to determine first inference information according to the reported configuration. The first inference information indicates the inference result for the reference signal resource sets configured for the first resource configuration and the P second resource configurations respectively. The inference result is obtained by inference based on the measurement result, which is obtained by measuring the reference signal resource set configured by the third resource configuration.

[0508] Optionally, the communication module 1520 is also used to send first inference information.

[0509] Optionally, the communication module 1520 is specifically used to send first inference information at a first reporting time, the first reporting time corresponding to a first reporting cycle.

[0510] Optionally, the communication module 1520 is further configured to send second inference information at a second reporting time, the second reporting time corresponding to a second reporting period, the second reporting period being different from the first reporting period, and the first reporting period being an integer multiple of the second reporting period. The second inference information may include inference results for reference signal resource sets configured for one or more of the P second resource configurations respectively.

[0511] Optionally, the communication module 1520 is further configured to transmit capability information, which indicates support for determining inference results for a first beam set and a plurality of fourth beam sets, the first beam set corresponding to a reference signal resource set configured in the first resource configuration, each of the plurality of fourth beam sets being a subset of the first beam set, and the reference signal resource set configured in one of the P second resource configurations corresponding to one of the plurality of fourth beam sets.

[0512] Optionally, the communication module 1520 is further configured to receive first information, which indicates a first reference signal resource set, the first reference signal resource set corresponding to a first beam set; the communication module 1520 is further configured to receive Q pieces of second information, which indicate Q fourth reference signal resource sets, each of the Q pieces of second information indicating one of the Q fourth reference signal resource sets, the Q fourth reference signal resource sets corresponding one-to-one with the Q fourth beam sets, where Q is a positive integer; wherein each of the Q fourth beam sets is a subset of the first beam set, the reference signal resource set configured in one of the P second resource configurations corresponds to one of the Q fourth beam sets, the first beam set corresponds to the reference signal resource set configured in the first resource configuration, and the first reference signal resource set and the Q fourth reference signal resource sets are used for training the first task.

[0513] Optionally, the processing module 1510 is further configured to perform training of the first task based on the first information, the second information, and the third measurement result, wherein the third measurement result is obtained based on measurements of a third set of reference signal resources and includes the input data corresponding to the training of the first task.

[0514] When the communication device 1500 is used to implement the function of the network device in the method embodiment shown in FIG8: the communication module 1520 is used to send a reported configuration, which is associated with multiple resource configurations. The multiple resource configurations include: a first resource configuration, P second resource configurations and a third resource configuration. Each of the multiple resource configurations is used to configure a set of reference signal resources. The P second resource configurations and the first resource configuration satisfy the following: the beam set corresponding to the reference signal resource set configured by each second resource configuration is a subset of the beam set corresponding to the reference signal resource set configured by the first resource configuration, where P is a positive integer.

[0515] Optionally, the communication module 1520 is also used to receive first inference information.

[0516] Optionally, the communication module 1520 is specifically used to receive first inference information at a first reporting time, the first reporting time corresponding to a first reporting cycle.

[0517] Optionally, the communication module 1520 is further configured to receive second inference information at a second reporting time, the second reporting time corresponding to a second reporting period, the second reporting period being different from the first reporting period, and the first reporting period being an integer multiple of the second reporting period. The second inference information may include inference results for reference signal resource sets configured for one or more of the P second resource configurations respectively.

[0518] Optionally, the communication module 1520 is further configured to receive capability information, which indicates support for determining inference results for a first beam set and a plurality of fourth beam sets, wherein the first beam set corresponds to a reference signal resource set configured in the first resource configuration, each of the plurality of fourth beam sets is a subset of the first beam set, and the reference signal resource set configured in one of the P second resource configurations corresponds to one of the plurality of fourth beam sets.

[0519] Optionally, the communication module 1520 is further configured to send first information, which indicates a first reference signal resource set, the first reference signal resource set corresponding to a first beam set; the communication module 1520 is further configured to send Q second information, which indicates Q fourth reference signal resource sets, each of the Q second information indicating one of the Q fourth reference signal resource sets, the Q fourth reference signal resource sets corresponding one-to-one with the Q fourth beam sets, where Q is a positive integer; wherein each of the Q fourth beam sets is a subset of the first beam set, the reference signal resource set configured in one of the P second resource configurations corresponds to one of the Q fourth beam sets, the first beam set corresponds to the reference signal resource set configured in the first resource configuration, and the first reference signal resource set and the Q fourth reference signal resource sets are used for training the first task.

[0520] In another alternative embodiment, the communication device 1500 is used to implement the functions of the terminal device or network device in the method embodiment shown in FIG14.

[0521] When the communication device 1500 is used to implement the functions of the terminal device in the method embodiment shown in FIG14: the communication module 1520 is used to receive first information and Q second information. The first information is used to indicate a first beam set, and the Q second information is used to indicate Q fourth beam sets, each second information indicating one fourth beam set, where Q is a positive integer; each of the Q fourth beam sets is a subset of the first beam set, and the first beam set and the Q fourth beam sets are used for training the first task.

[0522] Optionally, the communication module 1520 is further configured to receive third information, which indicates that the beam set corresponding to the output result of the first task includes a first beam set and Q fourth beam sets.

[0523] Optionally, the processing module 1510 is further configured to perform training of the first task based on the first information, the second information, and the third measurement result, wherein the third measurement result is obtained based on measurements of a third set of reference signal resources and includes the input data corresponding to the training of the first task.

[0524] Optionally, the processing module 1510 is specifically configured to: measure a first set of reference signal resources to obtain a first measurement result; measure Q sets of fourth reference signal resources to obtain Q second measurement results, wherein the Q sets of fourth reference signal resources correspond one-to-one with the Q sets of fourth beams, and the first measurement result and the Q second measurement results contain the label data corresponding to the training of the first task; obtain an inference result for the beam sets indicated by the first information and the second information respectively based on the third measurement result; and perform the training of the first task based on the label data and the inference result.

[0525] Optionally, the processing module 1510 is specifically configured to: measure the first reference signal resource set to obtain a first measurement result, the first measurement result including the label data corresponding to the training of the first task; obtain the inference result for the beam set indicated by the first information and the second information respectively based on the third measurement result; and perform the training of the first task based on the label data and the inference result.

[0526] Optionally, the communication module 1520 is also used to transmit capability information, which is used to instruct the first mission to support determining inference results for the first beam set and Q fourth beam sets.

[0527] Optionally, the communication module 1520 is further configured to receive fourth information, which indicates one or more beam sets, wherein any one of the one or more beam sets includes a first beam set and at least one of Q fourth beam sets, or includes at least two of Q fourth beam sets.

[0528] Optionally, the communication module 1520 is also used to receive sixth information; the sixth information includes Q association identifiers, each of which corresponds one-to-one with a Q fourth beam set; or, the sixth information includes one association identifier.

[0529] When the communication device 1500 is used to implement the function of the network device in the method embodiment shown in FIG14: the communication module 1520 is used to send a first message and Q second messages. The first message indicates a first beam set, and the Q second messages indicate Q fourth beam sets, each second message indicating one fourth beam set, where Q is a positive integer; each of the Q fourth beam sets is a subset of the first beam set, and the first beam set and the Q fourth beam sets are used for training the first task.

[0530] Optionally, the communication module 1520 is also configured to receive capability information, which is used to instruct the first mission to support determining inference results for the first beam set and Q fourth beam sets.

[0531] Optionally, the communication module 1520 is further configured to transmit fourth information, which indicates one or more beam sets, wherein any one of the one or more beam sets includes a first beam set and at least one of Q fourth beam sets, or includes at least two of Q fourth beam sets.

[0532] Optionally, the communication module 1520 is also used to transmit a sixth message; the sixth message includes Q association identifiers, each of which corresponds one-to-one with a Q fourth beam set; or, the sixth message includes a single association identifier.

[0533] A more detailed description of the processing module 1510 and the communication module 1520 described above can be found in the relevant descriptions in the method embodiments shown in Figures 8, 11 and 14.

[0534] It should be noted that the communication module can also be called a transceiver module, transceiver unit, transceiver, transceiver device, or transceiver apparatus, etc. The processing module can also be called a processor, processing board, processing unit, or processing apparatus, etc. Optionally, the communication module is used to execute the sending and receiving operations of the first or second communication device in the above method. The device in the communication module that implements the receiving function can be considered as the receiving module, and the device in the communication module that implements the sending function can be considered as the sending module; that is, the communication module can include both a receiving module and a sending module.

[0535] It should also be noted that, in one possible design, the aforementioned processing module and / or communication module can be implemented through virtual modules. For example, the processing module can be implemented through software functional units or virtual devices, and the communication module can be implemented through software functions or virtual devices. In another possible design, the processing module or communication module can also be implemented through physical devices. For example, if the device is implemented using a chip / chip circuit, the communication module can be an input / output circuit and / or a communication interface, performing input operations (corresponding to the aforementioned receiving operation) and output operations (corresponding to the aforementioned sending operation); the processing module can be an integrated processor, a microprocessor, or an integrated circuit.

[0536] The module division in this embodiment is illustrative and represents only one logical functional division; in actual implementation, other division methods may be used. Furthermore, the functional modules in the various examples of this embodiment can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0537] Figure 16 is a schematic diagram of a communication device provided in another embodiment of this application. As shown in Figure 16, the communication device 1600 includes a processing circuit 1610 and a communication circuit 1620. The processing circuit 1610 and the communication circuit 1620 are coupled to each other. The processing circuit can be one or more processors, or all or part of the circuitry in one or more processors used for control or processing functions. It is understood that when the communication device 1600 is a network device or a terminal device, the communication circuit 1620 can be a transceiver circuit, a transceiver, a communication interface, or an input / output interface. When the communication device 1600 is a chip for a network device or a terminal device, the communication circuit 1620 can be an input / output interface, a communication interface, or an input / output circuit. Optionally, the communication device 1600 may further include a memory 1630 for storing instructions executed by the processor, or storing input data required by the processor to execute instructions, or storing data generated after the processor executes instructions.

[0538] When the communication device 1600 is used to implement the methods shown in FIG8, FIG11 and FIG14, the processing circuit 1610 is used to implement the functions of the above-mentioned processing unit, and the communication circuit 1620 is used to implement the functions of the above-mentioned receiving unit and / or transmitting unit.

[0539] When the aforementioned communication device is a chip applied to a terminal device, the chip implements the functions of the terminal device in the above method embodiments. The chip can receive information from other modules (such as a radio frequency module or antenna) in the terminal device, information sent by the network device to the terminal device; or, the chip can send information to other modules (such as a radio frequency module or antenna) in the terminal device, information sent by the terminal device to the network device.

[0540] When the aforementioned communication device is a module applied to a network device (such as a base station), the module implements the functions of the network device in the above method embodiments. This module can receive information from other modules (such as radio frequency modules or antennas) in the network device, and this information may be sent from the terminal device to the network device; alternatively, the module can send information to other modules (such as radio frequency modules or antennas) in the network device, and this information may be sent from the network device to the terminal device. The module here can be the baseband chip of the base station, or a DU (Digital Unit) or other modules. The DU here can be a DU under an Open Radio Access Network (O-RAN) architecture.

[0541] It is understood that the processor in the embodiments of this application can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), graphics processing units (GPUs), neural processing units (NPUs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor can be a microprocessor or any conventional processor.

[0542] For example, the processor used for AI can be one or more of the following: graphics processing unit (GPU), neural processing unit (NPU), tensor processing unit (TPU), and data processing unit (DPU).

[0543] For example, one possible implementation of a processor for AI could be the AI ​​processor 1700 shown in Figure 17. Figure 17 is a schematic diagram of the structure of an AI processor provided in an embodiment of this application.

[0544] As shown in Figure 17, the AI ​​processor 1700 may include one or more of the following: an AI core, a digital vision pre-processing (DVPP) module, a task scheduler (TS), an L3 cache, an AI CPU, a control CPU, an L2 cache, a universal serial bus (USB) interface, a network interface card (NIC), a peripheral component interconnect express (PCIe) interface (PCIe is a high-speed serial computer expansion bus standard), a double data rate (DDR) / high bandwidth memory (HBM) interface, a general purpose input / output (GPIO) / inter-integrated circuit (I2C) bus, etc. It is understood that the specific meanings of these terms are well known to those skilled in the art and will not be elaborated upon here.

[0545] The terms “unit”, “module”, etc., used in this specification may be used to refer to computer-related entities, hardware, firmware, combinations of hardware and software, software, or software in execution.

[0546] The method steps in the embodiments of this application can be implemented in hardware or in software instructions executable by a processor. The software instructions can consist of corresponding software modules, which can be stored in random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disks, portable hard disks, compact disc read-only memory (CD-ROM), or any other form of storage medium well known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. The storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Alternatively, the ASIC can reside in a base station or terminal. The processor and storage medium can also exist as discrete components in the base station or terminal.

[0547] This application also provides a communication system, which includes the network device and terminal device described in the embodiments of this application. Optionally, it further includes an inference node and / or a training node, wherein the inference node is used to perform a first task, and the training node is used to perform training for the first task.

[0548] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that a computing device can store, or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital video disc (DVD)), or a semiconductor medium (e.g., a solid-state drive). The computer-readable storage medium includes instructions or program code that, when executed on a computing device, cause the computing device to perform the methods provided above.

[0549] This application also provides a computer program product, which may be a software or program product containing instructions capable of running on a computing device or stored on any usable medium. When the instructions are executed on the computing device, the computing device performs the methods provided above, or performs the functions of the apparatus provided above.

[0550] This application also provides a chip including at least one processor, which, when program instructions are executed by the at least one processor, causes the at least one processor to perform the methods executed by the terminal device or the network device in the above-described methods 800, 1100 or 1400.

[0551] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0552] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0553] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0554] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0555] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0556] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0557] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A communication method, characterized in that, include: The system receives a reported configuration, which is associated with multiple resource configurations. These multiple resource configurations include a first resource configuration, P second resource configurations, and a third resource configuration. Each of the multiple resource configurations is used to configure a set of reference signal resources. The P second resource configurations and the first resource configuration satisfy the following condition: the beam set corresponding to the reference signal resource set configured by each of the P second resource configurations is a subset of the beam set corresponding to the reference signal resource set configured by the first resource configuration, where P is a positive integer. Based on the reported configuration, first inference information is determined. The first inference information indicates the inference result for the reference signal resource sets configured for the first resource configuration and the P second resource configurations respectively. The inference result is obtained by inference based on measurement results, which are obtained by measuring the reference signal resource sets configured for the third resource configuration.

2. The method as described in claim 1, characterized in that, The set of reference signal resources configured in each of the P second resource configurations is a subset of the set of reference signal resources configured in the first resource configuration.

3. The method as described in claim 1, characterized in that, The reference signal resources configured in at least one of the P second resource configurations are not a subset of the reference signal resources configured in the first resource configuration, but correspond to the same set of beams.

4. The method according to any one of claims 1 to 3, characterized in that, The first inference information includes information about multiple beams, wherein the multiple beams are the complete set of the following terms: K beams in the first beam set, and K beams in the p-th second beam set obtained by iterating through p from 1 to P. p There are p-th beams, the first beam set corresponds to the reference signal resource set configured in the first resource configuration, and the p-th second beam set corresponds to the reference signal resource set configured in the p-th second resource configuration among the P second resource configurations, where p, K, and K are the reference signal resource sets configured in the first resource configuration. p It is a positive integer.

5. The method as described in claim 4, characterized in that, The reporting configuration also indicates one or more of the following: K, K1 to K P .

6. The method as described in claim 4, characterized in that, K, K1 to K P One or more of them are predefined.

7. The method according to any one of claims 1 to 3, characterized in that, The first inference information includes: a first inference result and P second inference results; wherein, the p-th second inference result among the P second inference results includes K in the p-th second beam set among the P second beam sets. p The information of each beam, wherein the p-th second beam set in the P second beam sets corresponds to the reference signal resource set configured in the p-th second resource configuration in the P second resource configurations; the first inference result includes information of K' beams in the first beam set, wherein the K' beams are the K' beams in the first beam set excluding the K” beams, and the K” beams are the complete set of the P second beam sets, wherein the first beam set corresponds to the reference signal resource set configured in the first resource configuration; K', K”, K p Both p and p are positive integers, with p taking values ​​from 1 to p.

8. The method as described in claim 7, characterized in that, The reporting configuration also indicates one or more of the following: K', K1 to K P .

9. The method as described in claim 7, characterized in that, K', K1 to K P One or more of them are predefined.

10. The method according to any one of claims 1 to 3, characterized in that, The first inference information includes: a first inference result and P second inference results; The first inference information includes K in the first beam set. sum Information about the beams, the first inference result includes (K) beams from the first beam set excluding the K”' beams. sum -K”') beams of information, wherein the first beam set corresponds to the reference signal resource set configured in the first resource configuration; the K”' beams are K from the p-th second beam set obtained by traversing p from 1 to P respectively. p The complete set of P beams, wherein the p-th second inference result among the P second inference results includes K in the p-th second beam set of the P second beam sets. p Information about each beam, wherein the p-th second beam set in the P second beam sets corresponds to the reference signal resource set configured in the p-th second resource configuration in the P second resource configurations, K p Greater than or equal to K p,min K p,min Less than K sum ,K”'、K sum K p K p,min Both p and p are positive integers, with p taking values ​​from 1 to p.

11. The method as described in claim 10, characterized in that, The reporting configuration also indicates one or more of the following: K sum and K 1,min To K P,min .

12. The method as described in claim 10, characterized in that, K sum and K 1,min To K P,min One or more of them are predefined.

13. The method according to any one of claims 1 to 12, characterized in that, The method further includes: Send the first inference information.

14. The method as described in claim 13, characterized in that, The reporting configuration also indicates a first reporting period, and the sending of the first inference information includes: The first inference information is sent at the first reporting time, which corresponds to the first reporting period.

15. The method as described in claim 14, characterized in that, The reporting configuration further indicates a second reporting period, which is different from the first reporting period, and the first reporting period is an integer multiple of the second reporting period; the method further includes: The second inference information is sent at the second reporting time, which corresponds to the second reporting period. The second inference information includes the inference results of the reference signal resource sets configured for one or more of the P second resource configurations.

16. The method according to any one of claims 1 to 15, characterized in that, Prior to receiving the reported configuration, the method further includes: Send capability information, which indicates support for determining inference results for a first beam set and multiple fourth beam sets, wherein the first beam set corresponds to the reference signal resource set configured by the first resource configuration, each of the multiple fourth beam sets is a subset of the first beam set, and the reference signal resource set configured by one of the P second resource configurations corresponds to one of the multiple fourth beam sets.

17. The method according to any one of claims 1 to 16, characterized in that, Prior to receiving the reported configuration, the method further includes: Receive first information, the first information being used to indicate a first reference signal resource set, the first reference signal resource set corresponding to a first beam set; Receive Q second pieces of information, the Q second pieces of information are used to indicate Q sets of fourth reference signal resources, each of the Q second pieces of information is used to indicate one of the reference signal resource sets in the Q sets of fourth reference signal resources, the Q sets of fourth reference signal resources correspond one-to-one with the Q sets of fourth beams, and Q is a positive integer; Wherein, each of the Q fourth beam sets is a subset of the first beam set, the reference signal resource set configured in one of the P second resource configurations corresponds to one of the Q fourth beam sets, the first beam set corresponds to the reference signal resource set configured in the first resource configuration, and the first reference signal resource set and the Q fourth reference signal resource sets are used for training the first task.

18. The method as described in claim 17, characterized in that, The method further includes: Based on the first information, the second information, and the third measurement result, the training of the first task is performed. The third measurement result is obtained by measuring a third set of reference signal resources and includes the input data corresponding to the training of the first task.

19. A communication method, characterized in that, include: Send a reporting configuration, which is associated with multiple resource configurations. The multiple resource configurations include: a first resource configuration, P second resource configurations, and a third resource configuration. Each of the multiple resource configurations is used to configure a set of reference signal resources. The P second resource configurations have a first relationship with the first resource configuration, where P is a positive integer.

20. The method as described in claim 19, characterized in that, The set of reference signal resources configured in each of the P second resource configurations is a subset of the set of reference signal resources configured in the first resource configuration.

21. The method as described in claim 19, characterized in that, The reference signal resources configured in at least one of the P second resource configurations are not a subset of the reference signal resources configured in the first resource configuration, but correspond to the same set of beams.

22. The method according to any one of claims 19 to 21, characterized in that, The method further includes: Receive first inference information, which indicates the inference results for reference signal resources configured for the first resource configuration and the P second resource configurations respectively. The inference results are obtained by inference based on measurement results, which are obtained by measurement based on the set of reference signal resources configured for the third resource configuration.

23. The method as described in claim 22, characterized in that, The first inference information includes information about multiple beams, wherein the multiple beams are the complete set of the following terms: K beams in the first beam set, and K beams in the p-th second beam set obtained by iterating through p from 1 to P. p There are p-th beams, the first beam set corresponds to the reference signal resource set configured in the first resource configuration, and the p-th second beam set corresponds to the reference signal resource set configured in the p-th second resource configuration among the P second resource configurations, where p, K, and K are the reference signal resource sets configured in the first resource configuration. p It is a positive integer.

24. The method as described in claim 23, characterized in that, The reporting configuration also indicates one or more of the following: K, K1 to K P .

25. The method as described in claim 23, characterized in that, K, K1 to K P One or more of them are predefined.

26. The method as described in claim 22, characterized in that, The first inference information includes: a first inference result and P second inference results; wherein, the p-th second inference result among the P second inference results includes K in the p-th second beam set among the P second beam sets. p The information of each beam, wherein the p-th second beam set in the P second beam sets corresponds to the reference signal resource set configured in the p-th second resource configuration in the P second resource configurations; the first inference result includes information of K' beams in the first beam set, wherein the K' beams are the K' beams in the first beam set excluding the K” beams, and the K” beams are the complete set of the P second beam sets, wherein the first beam set corresponds to the reference signal resource set configured in the first resource configuration; K', K”, K p Both p and p are positive integers, with p taking values ​​from 1 to p.

27. The method as described in claim 26, characterized in that, The reporting configuration also indicates one or more of the following: K', K1 to K P .

28. The method as described in claim 26, characterized in that, K', K1 to K P One or more of them are predefined.

29. The method as described in claim 22, characterized in that, The first inference information includes: a first inference result and P second inference results; The first inference information includes K in the first beam set. sum Information about the beams, the first inference result includes (K) beams from the first beam set excluding the K”' beams. sum -K”') beams of information, wherein the first beam set corresponds to the reference signal resource set configured in the first resource configuration; the K”' beams are K from the p-th second beam set obtained by traversing p from 1 to P respectively. p The complete set of P beams, wherein the p-th second inference result among the P second inference results includes K in the p-th second beam set of the P second beam sets. p Information about each beam, wherein the p-th second beam set in the P second beam sets corresponds to the reference signal resource set configured in the p-th second resource configuration in the P second resource configurations, K p Greater than or equal to K p,min K p,min Less than K sum ,K”'、K sum K p K p,min Both p and p are positive integers, with p taking values ​​from 1 to p.

30. The method as described in claim 29, characterized in that, The reporting configuration also indicates one or more of the following: K sum and K 1,min To K P,min .

31. The method as described in claim 29, characterized in that, K sum and K 1,min To K P,min One or more of them are predefined.

32. The method according to any one of claims 22 to 31, characterized in that, The reporting configuration also indicates a first reporting period, and the receiving of the first inference information includes: The first inference information is received at the first reporting time, which corresponds to the first reporting period.

33. The method as described in claim 32, characterized in that, The reporting configuration further indicates a second reporting period, which is different from the first reporting period, and the first reporting period is an integer multiple of the second reporting period; the method further includes: Second inference information is received at a second reporting time, which corresponds to a second reporting period. The second inference information includes inference results for reference signal resource sets configured for one or more of the P second resource configurations.

34. The method according to any one of claims 19 to 33, characterized in that, Before sending the reporting configuration, the method further includes: The capability information is used to indicate support for determining inference results for a first beam set and a plurality of fourth beam sets. The first beam set corresponds to the reference signal resource set configured by the first resource configuration. Each of the plurality of fourth beam sets is a subset of the first beam set. The reference signal resource set configured by one of the P second resource configurations corresponds to one of the fourth beam sets in the plurality of fourth beam sets.

35. The method according to any one of claims 19 to 34, characterized in that, Before sending the reporting configuration, the method further includes: Send first information, the first information being used to indicate a first reference signal resource set, the first reference signal resource set corresponding to a first beam set; Send Q second information messages, which are used to indicate Q fourth reference signal resource sets. Each of the Q second information messages is used to indicate one of the reference signal resource sets in the Q fourth reference signal resource sets. The Q fourth reference signal resource sets correspond one-to-one with the Q fourth beam sets, and Q is a positive integer. Wherein, each of the Q fourth beam sets is a subset of the first beam set, the reference signal resource set configured in one of the P second resource configurations corresponds to one of the Q fourth beam sets, the first beam set corresponds to the reference signal resource set configured in the first resource configuration, and the first reference signal resource set and the Q fourth reference signal resource sets are used for training the first task.

36. The method according to any one of claims 1 to 35, characterized in that, The reporting configuration is used to configure Channel State Information (CSI) reports. The amount of processing resources occupied by the CSI reports is related to the number of reference signal resources in the reference signal resource set configured by the first resource configuration.

37. The method as described in claim 36, characterized in that, The start time of the CSI report's occupation of the processing resources is the time of the last reference signal received before the CSI reference resource, and the end time is the reporting time of the CSI report.

38. A communication method, characterized in that, include: Receive first information, which is used to indicate a first beam set; Receive Q pieces of second information, wherein the Q pieces of second information are used to indicate Q sets of fourth beams, and each of the Q pieces of second information is used to indicate one of the Q sets of fourth beams, where Q is a positive integer; Each of the Q fourth beam sets is a subset of the first beam set, and the first beam set and the Q fourth beam sets are used for training the first task.

39. The method as described in claim 38, characterized in that, The first information is used to configure a first reference signal resource set, which includes X reference signal resources and a first beam set, which includes X beams. The X reference signal resources correspond one-to-one with the X beams, and X is a positive integer.

40. The method as described in claim 39, characterized in that, The Q pieces of second information are used to configure Q sets of fourth reference signal resources, and the Q sets of fourth reference signal resources correspond one-to-one with the Q sets of fourth beams; wherein... Each of the Q fourth reference signal resource sets is a subset of the first reference signal resource set; or, The q-th fourth reference signal resource set in the Q fourth reference signal resource sets includes Y. q A reference signal resource, the Y q One reference signal resource and Y in the X reference signal resources q Each reference signal resource has a one-to-one mapping relationship, and the two reference signal resources with the mapping relationship have the same beam.

41. The method as described in claim 38, characterized in that, Each of the Q second pieces of information includes a bit map, the bit map includes X indicator bits, the X indicator bits correspond one-to-one with the X beams, and each of the X indicator bits in the qth second piece of information is used to indicate whether the corresponding beam belongs to the fourth beam set indicated by the qth second piece of information.

42. The method according to any one of claims 38 to 41, characterized in that, The method further includes: Receive third information, which indicates that the beam set corresponding to the output result of the first task includes the first beam set and the Q fourth beam sets.

43. The method as described in claim 42, characterized in that, The method further includes: Based on the first information, the second information, and the third measurement result, the training of the first task is performed. The third measurement result is obtained by measuring a third set of reference signal resources and includes the input data corresponding to the training of the first task.

44. The method as described in claim 43, characterized in that, The step of basing information on the first information, the second information, and the third measurement result includes: The first reference signal resource set is measured to obtain a first measurement result; Measurements are performed on Q sets of fourth reference signal resources to obtain Q second measurement results. The Q sets of fourth reference signal resources correspond one-to-one with the Q sets of fourth beams. The first measurement results and the Q second measurement results contain the label data corresponding to the training of the first task. Based on the third measurement result, inference results are obtained for the beam sets indicated by the first information and the second information, respectively; Training for the first task is performed based on the labeled data and the inference results.

45. The method as described in claim 43, characterized in that, The step of basing information on the first information, the second information, and the third measurement result includes: The first reference signal resource set is measured to obtain a first measurement result, which includes the label data corresponding to the training of the first task; Based on the third measurement result, inference results are obtained for the beam sets indicated by the first information and the second information, respectively; Training for the first task is performed based on the labeled data and the inference results.

46. ​​The method according to any one of claims 38 to 45, characterized in that, The method further includes: Send capability information, which is used to indicate that the first task supports determining inference results for the first beam set and the Q fourth beam sets.

47. The method according to any one of claims 38 to 46, characterized in that, The method further includes: Receive fourth information, the fourth information being used to indicate one or more beam set groups, any one of the one or more beam set groups including the first beam set and at least one of the Q fourth beam sets, or including at least two of the Q fourth beam sets.

48. The method according to any one of claims 38 to 47, characterized in that, The method further includes: The fifth information is received, which is also used to indicate the performance requirements of the output results of the first task.

49. The method according to any one of claims 38 to 48, characterized in that, The Q sets of fourth beams may have the same or different association identifiers as the first beam set.

50. The method as described in claim 49, characterized in that, The method further includes: Receive the sixth information; the sixth information includes Q association identifiers, each of which corresponds one-to-one with one of the Q fourth beam sets; or, the sixth information includes one association identifier.

51. A communication method, characterized in that, include: Send a first message, which is used to indicate a first beam set; Send Q second messages, the Q second messages being used to indicate Q fourth beam sets, each of the Q second messages being used to indicate one of the Q fourth beam sets, where Q is a positive integer; Each of the Q sets of fourth beams is a subset of the first beam set, and the first beam set and the Q sets of fourth beams are used for training the first task.

52. The method as described in claim 51, characterized in that, The first information is used to configure a first reference signal resource set, which includes X reference signal resources and a first beam set, which includes X beams. The X reference signal resources correspond one-to-one with the X beams, and X is a positive integer.

53. The method as described in claim 52, characterized in that, The Q pieces of second information are used to configure Q sets of fourth reference signal resources, and the Q sets of fourth reference signal resources correspond one-to-one with the Q sets of fourth beams; wherein... Each of the Q fourth reference signal resource sets is a subset of the first reference signal resource set; or, The q-th fourth reference signal resource set in the Q fourth reference signal resource sets includes Y. q A reference signal resource, the Y q One reference signal resource and Y in the X reference signal resources q Each reference signal resource has a one-to-one mapping relationship, and the two reference signal resources with the mapping relationship have the same beam.

54. The method as described in claim 51, characterized in that, Each of the Q second pieces of information includes a bit map, the bit map includes X indicator bits, the X indicator bits correspond one-to-one with the X beams, and each of the X indicator bits in the qth second piece of information is used to indicate whether the corresponding beam belongs to the fourth beam set indicated by the qth second piece of information.

55. The method according to any one of claims 51 to 54, characterized in that, The method further includes: Send a third message, which indicates that the beam set corresponding to the output result of the first task includes the first beam set and the Q fourth beam sets.

56. The method according to any one of claims 51 to 55, characterized in that, The method further includes: Receive capability information, which is used to indicate that the first task supports determining inference results for the first beam set and the Q fourth beam sets.

57. The method according to any one of claims 51 to 56, characterized in that, The method further includes: A fourth message is sent, the fourth message indicating one or more beam set groups, any one of the one or more beam set groups including the first beam set and at least one of the Q fourth beam sets, or including at least two of the Q fourth beam sets.

58. The method according to any one of claims 51 to 57, characterized in that, The method further includes: A fifth message is sent, which is also used to indicate the performance requirements of the output of the first task.

59. The method according to any one of claims 51 to 58, characterized in that, The Q sets of fourth beams may have the same or different association identifiers as the first beam set.

60. The method as described in claim 59, characterized in that, The method further includes: Send a sixth message; the sixth message includes Q association identifiers, each of which corresponds one-to-one with one of the Q fourth beam sets; or, the sixth message includes one association identifier.

61. A communication device, characterized in that, It includes modules or units for implementing the method as described in any one of claims 1 to 37, or modules or units for implementing the method as described in any one of claims 38 to 60.

62. A communication device, characterized in that, The device includes one or more processors and communication circuitry, the communication circuitry being used for at least one of inputting or outputting signals; the one or more processors are used to implement the method as described in any one of claims 1 to 37, or to implement the method as described in any one of claims 38 to 60.

63. A communication system, characterized in that, It includes a first device and a second device; wherein, The first device is used to perform the method as described in any one of claims 1 to 18, or any one of claims 36 and 37 when referring to any one of claims 1 to 18; The second device is used to perform the method as claimed in any one of claims 19 to 35, or any one of claims 36 and 37 when any one of claims 19 to 35 is invoked.

64. A communication system, characterized in that, Including a third device and a fourth device; wherein, The third device is used to perform the method as described in any one of claims 38 to 50, and the fourth device is used to perform the method as described in any one of claims 51 to 60.

65. A readable storage medium, characterized in that, Used to store programs or instructions that, when executed, are implemented in accordance with the method of any one of claims 1 to 37, or in accordance with any one of claims 38 to 60.

66. A computer program product, characterized in that, Includes a program or instructions that, when executed, implement the method as claimed in any one of claims 1 to 37, or implement the method as claimed in any one of claims 38 to 60.

Images