Communication method and device

Abstract

An embodiment of this application provides a communication method and apparatus for reducing redundant overhead in a measurement process and improving communication efficiency. In this method, after a terminal obtains RS measurement results through measurements by using a first RS resource set, the terminal may perform prediction on the RS measurement results by using a neural network model to determine a candidate RS resource set belonging to a second RS resource set. In this case, the terminal may further select a target RS resource set from the candidate RS resource sets as the RS resource to be finally reported, thereby reducing redundant overhead in the measurement process and improving communication efficiency.

Description

[Technical Field]

[0001] [CROSS-REFERENCE TO RELATED APPLICATIONS] This application claims priority to Chinese Patent Application No. 202310160189.X, entitled "COMMUNICATION METHOD AND APPARATUS," filed with the State Intellectual Property Office of the People's Republic of China on February 16, 2023, which is incorporated herein by reference in its entirety.

[0002] [Technical field] This application relates to the field of wireless communications, and more particularly to communication methods and devices. [Background technology]

[0003] The 3rd Generation Partnership Project (3GPP) protocol introduces a signal transmission mechanism based on beamforming technology. This transmission mechanism requires beam management to achieve alignment between transmit and receive beams. For example, a network device may transmit a reference signal to a terminal at multiple times by using different beams. The terminal may determine an optimal beam, e.g., an optimal transmit beam and / or an optimal receive beam, by measuring the reference signal. Transmitting and receiving a reference signal requires high air interface overhead. Therefore, a possible solution is for the terminal to preconfigure the optimal beam through a neural network. For example, the network device may transmit a reference signal by using only a small number of beams. The terminal may obtain measurement results by measuring the reference signal and input the measurement results into a neural network to predict the optimal possible beam. In this way, the number of measurements can be reduced, thereby reducing air interface overhead.

[0004] However, the pre-configured optimal beam of the neural network may differ from the actual optimal beam, resulting in redundant overhead in the measurement process, which in turn affects communication efficiency. Summary of the Invention

[0005] The embodiments of this application provide a communication method and apparatus for reducing redundant overhead in the measurement process and improving communication efficiency.

[0006] In order to achieve the above objectives, the following technical solutions are used in the embodiments of this application.

[0007] According to a first aspect, a communication method is provided. The method includes: a terminal measures RSs from a network device by using a first reference signal RS resource set to obtain RS measurement results, and inputs the measurement results into a neural network model to determine a target RS resource set. The target RS resource set is determined from candidate RS resource sets, and the candidate RS resource set is determined by the neural network model based on the measurement results, and the candidate RS resource set belongs to a second RS resource set. The RSs transmitted by using the first RS resource set are the same as or different from the RSs transmitted by using the second RS resource set.

[0008] It can be seen from the method according to the first aspect that after the terminal obtains the measurement result of the RS through the measurement by using the first RS resource set, the terminal may perform prediction on the measurement result of the RS by using a neural network model to determine a candidate RS resource set belonging to the second RS resource set, in which case the terminal may further select a target RS resource set from the candidate RS resource sets as the RS resource to be finally reported, thereby reducing redundant overhead in the measurement process and improving communication efficiency.

[0009] In a possible design solution, the target RS resource set is determined from the candidate RS resource sets based on a first threshold and information about the candidate RS resource sets. The first threshold is related to the information about the candidate RS resource sets. For example, the first threshold may be set based on the information type of the candidate RS resource sets (see the related description below for details), so that redundant RSs in the candidate RS resource sets can be more accurately eliminated.

[0010] Optionally, the candidate RS resource set is determined by a neural network model based on RS measurement results, specifically as follows: Information about the candidate RS resource set is determined by a neural network model based on RS measurement results. The information about the candidate RS resource set may be used to indicate that the RS resources in the candidate RS resource set can be used as optimal RS resources, or the likelihood that the RS resources in the candidate RS resource set will be used as optimal RS resources. In other words, possible optimal RS resources may be predicted by using a neural network model to reduce beam management overhead.

[0011] Optionally, the information about the candidate RS resource sets may include at least one of the following: a probability that each RS resource in the candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the candidate RS resource set, or an angle of each RS resource in the candidate RS resource set, where the angle of each RS resource in the candidate RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0012] It can be seen from the above description that in some cases, the probability of an optimal RS resource may explicitly indicate the likelihood that an RS resource is used as an optimal RS resource, and the network device can perform effective resource scheduling or configuration based on the probability to avoid redundant overhead. In some other cases, the terminal needs to report the signal quality of the RS resource, e.g., the RSRP, to the network device. Thus, the signal quality of the RS resource is reused to implicitly indicate that the RS resource may be used as an optimal RS resource, thereby reducing reporting overhead. In other cases, the angle of the RS resource may also explicitly indicate the likelihood that the RS resource is used as an optimal RS resource. In this way, the network device can also perform effective resource scheduling or configuration based on the angle to avoid redundant overhead.

[0013] Furthermore, the information about the target RS resource set and the first threshold satisfy at least one of the following relationships: the probability that an RS resource in the target RS resource set is an optimal RS resource is greater than the probability represented by the first threshold, the signal quality of the RS resource in the target RS resource set is greater than the signal quality represented by the first threshold, or the angle of each RS resource in the RS resource set is less than the angle represented by the first threshold. In other words, the RS resource that is most likely to be used as the optimal RS resource can be selected from the candidate RS resource set based on the first threshold, avoiding redundant overhead.

[0014] Optionally, the method according to the first aspect may further include: the terminal receives instruction information from the network device, the instruction information indicating a first threshold value. In other words, the first threshold value may be flexibly configured by the network side according to actual requirements.

[0015] Optionally, the first threshold may alternatively be pre-configured in the neural network model, or may be pre-configured locally on the terminal to avoid unnecessary overhead caused by the network side separately configuring the first threshold.

[0016] In a possible design solution, the method according to the first aspect may further include: the terminal transmits information about the target RS resource set to the network device, so that the network device determines which RS resources can be used as optimal RS resources. For example, the information about the target RS resource set includes at least one of the following: identifiers of the RS resources in the target RS resource set, signal qualities of the RS resources in the target RS resource set, or angles of each RS resource in the RS resource set. The angle of each RS resource in the target RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource. Specifically, in some cases, the information about the target RS resource set may include identifiers and signal qualities of the RS resources to jointly indicate which RS resources can be used as optimal RS resources. Alternatively, in other cases, the information about the target RS resource set may include the angles of the RS resources. In this way, the network device may also determine that RS resources can be used as optimal RS resources based only on the angles of the RS resources.

[0017] In a possible design solution, there may be multiple target RS resource sets, which may be separately configured for use in different time units to ensure that optimal RS resources are available for air interface transmission in each time unit, thereby ensuring the stability and reliability of air interface transmission.

[0018] According to a second aspect, a communication method is provided, the method including: a network device transmits an RS to a terminal by using a first reference signal RS resource set, receives measurement results of the RS fed back by the terminal, and inputs the measurement results into a neural network model to determine a target RS resource set, the target RS resource set is determined from candidate RS resource sets, the candidate RS resource set is determined by the neural network model based on the measurement results, the candidate RS resource set belongs to a second RS resource set, and the RS transmitted by using the first RS resource set is the same as or different from the RS transmitted by using the second RS resource set.

[0019] It can be seen from the method according to the second aspect that after the terminal obtains the measurement result of the RS through measurement by using the first RS resource set, the terminal may directly report the measurement result to the network device. In this way, the network device may perform prediction on the measurement result of the RS by using a neural network model to determine candidate RS resource sets belonging to the second RS resource set, and further select a target RS resource set from the candidate RS resource sets as the RS resource to be finally reported, thereby reducing redundant overhead in the measurement process and improving communication efficiency.

[0020] In a possible design solution, the target RS resource set is determined from the candidate RS resource sets based on a first threshold and information about the candidate RS resource sets, where the first threshold is related to the information about the candidate RS resource sets.

[0021] Optionally, the candidate RS resource set is determined by a neural network model based on the measurement result, specifically as follows: Information about the candidate RS resource set is determined by a neural network model based on the measurement result.

[0022] Furthermore, the information about the candidate RS resource sets includes at least one of the following: a probability that each RS resource in the candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the candidate RS resource set, or an angle of each RS resource in the candidate RS resource set, where the angle of each RS resource in the candidate RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0023] Furthermore, the information about the target RS resource set and the first threshold satisfy at least one of the following relationships: the probability that the RS resources in the target RS resource set are optimal RS resources is greater than the probability represented by the first threshold, the signal quality of the RS resources in the target RS resource set is greater than the signal quality represented by the first threshold, or the angle of each RS resource in the RS resource set is less than the angle represented by the first threshold.

[0024] Optionally, the first threshold may be pre-configured in the neural network model.

[0025] In a possible design solution, the method according to the second aspect may further include: the network device sends a target RS resource set to the terminal.

[0026] In a possible design solution, there are multiple target RS resource sets, and the multiple target RS resource sets are configured for use in different time units.

[0027] In a possible design solution, the measurement results of the RS are all or a part of the measurement results, in other words, the terminal may choose to report only a part of the measurement results to reduce overhead.

[0028] Furthermore, the technical effects of the method according to the second aspect refer to the technical effects of the method according to the first aspect, and the details will not be described again herein.

[0029] According to a third aspect, a communication method is provided. The method includes: a terminal measures RSs from a network device by using a first reference signal RS resource set to obtain RS measurement results, and inputs the measurement results into a neural network model to determine a target RS resource set group. The target RS resource set group includes a first target RS resource set corresponding to a first time unit and a second target RS resource set corresponding to a second time unit, where the first time unit is different from the second time unit. The first target RS resource set is determined from a first candidate RS resource set, and the second target RS resource set is determined from the second candidate RS resource set. The first candidate RS resource set and the second candidate RS resource set are determined by the neural network model based on the measurement results, where both the first candidate RS resource set and the second candidate RS resource set belong to the second RS resource set. The RS transmitted by using the first RS resource set is the same as or different from the RS transmitted by using the second RS resource set.

[0030] In a possible design solution, a first target RS resource set corresponding to a first time unit means that the first target RS resource set is configured for use in the first time unit.

[0031] In a possible design solution, the second target RS resource set corresponding to the second time unit means that the second target RS resource set is configured for use in the second time unit.

[0032] In a possible design solution, a first target RS resource set is determined from a first candidate RS resource set based on a first threshold and information about the first candidate RS resource set, where the first threshold is related to the information about the first candidate RS resource set.

[0033] Optionally, the first candidate RS resource set is determined by a neural network model based on measurement results of RSs, specifically as follows: Information about the first candidate RS resource set is determined by a neural network model based on measurement results of RSs.

[0034] Optionally, the information about the first candidate RS resource set includes at least one of the following: a probability that each RS resource in the first candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the first candidate RS resource set, or an angle of each RS resource in the first candidate RS resource set, where the angle of each RS resource in the first candidate RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0035] Furthermore, the information about the first target RS resource set and the first threshold satisfy at least one of the following relationships: the probability that the RS resources in the first target RS resource set are optimal RS resources is greater than the probability represented by the first threshold, the signal quality of the RS resources in the first target RS resource set is greater than the signal quality represented by the first threshold, or the angle of each RS resource in the first RS resource set is less than the angle represented by the first threshold.

[0036] Optionally, the method according to the third aspect may further include: the terminal receives first indication information from the network device, where the first indication information indicates a first threshold value.

[0037] Optionally, the first threshold may alternatively be pre-configured in the neural network model or may be pre-configured locally at the terminal.

[0038] In a possible design solution, the method according to the third aspect may further include: the terminal transmits information about the first target RS resource set to the network device. For example, the information about the first target RS resource set includes at least one of the following: identifiers of RS resources in the first target RS resource set, signal qualities of RS resources in the first target RS resource set, or angles of each RS resource in the first RS resource set. The angle of each RS resource in the first target RS resource set is an angle difference between a transmit beam and a receive beam corresponding to the RS resource.

[0039] In a possible design solution, the second target RS resource set is determined from the second candidate RS resource set based on a second threshold and information about the second candidate RS resource set, where the second threshold is related to the information about the second candidate RS resource set.

[0040] Optionally, the second candidate RS resource set is determined by a neural network model based on the measurement results of the RSs, specifically as follows: Information about the second candidate RS resource set is determined by a neural network model based on the measurement results of the RSs.

[0041] Optionally, the information about the second candidate RS resource set includes at least one of the following: a probability that each RS resource in the second candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the second candidate RS resource set, or an angle of each RS resource in the second candidate RS resource set, where the angle of each RS resource in the candidate RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0042] Optionally, the method according to the third aspect may further include: the terminal receives second indication information from the network device, where the second indication information indicates a second threshold value.

[0043] Optionally, the second threshold may alternatively be pre-configured in the neural network model or may be pre-configured locally at the terminal.

[0044] In a possible design solution, the method according to the third aspect may further include: the terminal transmits information about the second target RS resource set to the network device. For example, the information about the second target RS resource set includes at least one of the following: identifiers of RS resources in the second target RS resource set, signal qualities of RS resources in the second target RS resource set, or angles of each RS resource in the second RS resource set. The angle of each RS resource in the second target RS resource set is an angle difference between a transmit beam and a receive beam corresponding to the RS resource.

[0045] Furthermore, the technical effects of the method according to the third aspect refer to the technical effects of the method according to the first aspect, and the details will not be described again herein.

[0046] According to a fourth aspect, a communication method is provided. The method includes: a network device transmits an RS to a terminal by using a first reference signal RS resource set and receives measurement results of the RS fed back by the terminal; and the network device inputs the measurement results into a neural network model to determine a target RS resource set group. The target RS resource set group includes a first target RS resource set corresponding to a first time unit and a second target RS resource set corresponding to a second time unit, where the first time unit is different from the second time unit. The first target RS resource set is determined from a first candidate RS resource set, and the second target RS resource set is determined from the second candidate RS resource set. The first candidate RS resource set and the second candidate RS resource set are determined by the neural network model based on the measurement results, where both the first candidate RS resource set and the second candidate RS resource set belong to the second RS resource set. The RS transmitted by using the first RS resource set is the same as or different from the RS transmitted by using the second RS resource set.

[0047] In a possible design solution, a first target RS resource set corresponding to a first time unit means that the first target RS resource set is configured for use in the first time unit.

[0048] In a possible design solution, the second target RS resource set corresponding to the second time unit means that the second target RS resource set is configured for use in the second time unit.

[0049] In a possible design solution, a first target RS resource set is determined from a first candidate RS resource set based on a first threshold and information about the first candidate RS resource set, where the first threshold is related to the information about the first candidate RS resource set.

[0050] Optionally, the first candidate RS resource set is determined by a neural network model based on measurement results of RSs, specifically as follows: Information about the first candidate RS resource set is determined by a neural network model based on measurement results of RSs.

[0051] Furthermore, the information about the first candidate RS resource set includes at least one of the following: a probability that each RS resource in the first candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the first candidate RS resource set, or an angle of each RS resource in the first candidate RS resource set, where the angle of each RS resource in the first candidate RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0052] Furthermore, the information about the first target RS resource set and the first threshold satisfy at least one of the following relationships: the probability that the RS resources in the first target RS resource set are optimal RS resources is greater than the probability represented by the first threshold, the signal quality of the RS resources in the first target RS resource set is greater than the signal quality represented by the first threshold, or the angle of each RS resource in the first RS resource set is less than the angle represented by the first threshold.

[0053] Optionally, the first threshold may be pre-configured in the neural network model.

[0054] In a possible design solution, the method according to the second aspect may further include: the network device sends a first target RS resource set to the terminal.

[0055] In a possible design solution, the second target RS resource set is determined from the second candidate RS resource set based on a second threshold and information about the second candidate RS resource set, where the second threshold is related to the information about the second candidate RS resource set.

[0056] Optionally, the second candidate RS resource set is determined by a neural network model based on the measurement results of the RSs, specifically as follows: Information about the second candidate RS resource set is determined by a neural network model based on the measurement results of the RSs.

[0057] Furthermore, the information about the second candidate RS resource set includes at least one of the following: a probability that each RS resource in the second candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the second candidate RS resource set, or an angle of each RS resource in the second candidate RS resource set, where the angle of each RS resource in the second candidate RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0058] Furthermore, the information about the second target RS resource set and the second threshold satisfy at least one of the following relationships: the probability that the RS resources in the second target RS resource set are optimal RS resources is greater than the probability represented by the second threshold, the signal quality of the RS resources in the second target RS resource set is greater than the signal quality represented by the second threshold, or the angle of each RS resource in the second RS resource set is less than the angle represented by the second threshold.

[0059] Optionally, the second threshold may be pre-configured in the neural network model.

[0060] In a possible design solution, the method according to the second aspect may further include: the network device sends a second target RS resource set to the terminal.

[0061] In a possible design solution, the RS measurement is the entire measurement or a part of the measurement.

[0062] Furthermore, the technical effects of the method according to the fourth aspect refer to the technical effects of the method according to the second aspect, and the details will not be described again herein.

[0063] According to a fifth aspect, there is provided a communications device, the communications device including modules configured to perform the method according to the first or third aspect, such as a transceiver module and a processing module, for example, the transceiver module representing the receiving and transmitting functions of the communications device, and the processing module configured to perform functions of the communications device other than the receiving and transmitting functions.

[0064] Optionally, the transceiver module may include a transmitting module and a receiving module, wherein the transmitting module is configured to implement a transmitting function of the communication device according to the fifth aspect, and the receiving module is configured to implement a receiving function of the communication device according to the fifth aspect.

[0065] Optionally, the communication device according to the fifth aspect may further include a storage module for storing a program or instructions, which, when executed by the processing module, may enable the communication device to perform the method according to the first or third aspect.

[0066] It may be understood that the communication device according to the fifth aspect may be a terminal, or a chip (system) or other part or component that may be disposed in a terminal, or may be a device that includes a terminal, which is not limited in this application.

[0067] Furthermore, the technical effects of the communication device according to the fifth aspect refer to the technical effects of the method according to the first aspect, and the details will not be described again in this specification.

[0068] According to a sixth aspect, there is provided a communications device, the communications device including modules configured to perform the method according to the second or fourth aspect, such as a transceiver module and a processing module, for example, the transceiver module representing the receiving and transmitting functions of the communications device, and the processing module configured to perform functions of the communications device other than the receiving and transmitting functions.

[0069] Optionally, the transceiver module may include a transmitting module and a receiving module, wherein the transmitting module is configured to implement a transmitting function of the communication device according to the sixth aspect, and the receiving module is configured to implement a receiving function of the communication device according to the sixth aspect.

[0070] Optionally, the communication device according to the sixth aspect may further include a storage module for storing a program or instructions, which, when executed by the processing module, may enable the communication device to perform the method according to the second or fourth aspect.

[0071] It can be understood that the communication device according to the sixth aspect may be a network, or a chip (system) or other part or component that may be disposed in a terminal, or may be a device that includes a terminal, which is not limited in this application.

[0072] Furthermore, the technical effects of the communication device according to the sixth aspect refer to the technical effects of the method according to the second aspect, and the details will not be described again in this specification.

[0073] According to a seventh aspect, there is provided a communication device, the communication device including a processor, the processor being configured to execute a method according to any one of the possible implementation manners of the first to fourth aspects.

[0074] In a possible design solution, the communication device according to the seventh aspect may further include a transceiver. The transceiver may be a transceiver circuit or an interface circuit. The transceiver may be used by the communication device according to the seventh aspect to communicate with other communication devices.

[0075] In a possible design solution, the communication device according to the seventh aspect may further include a memory. The memory may be integrated with the processor or may be located separately. The memory may be configured to store computer programs and / or data related to the method according to any one of the first to fourth aspects.

[0076] In an embodiment of this application, the communication device according to the seventh aspect may be a device according to any one of the first to fourth aspects, or a chip (system) or other part or component that may be disposed within the device, or an apparatus including the device.

[0077] Furthermore, for the technical effects of the communication device according to the seventh aspect, refer to the technical effects of the method according to any one of the implementation methods of the first to fourth aspects, and the details will not be described again in this specification.

[0078] According to an eighth aspect, there is provided a communication device, the communication device including a processor coupled to a memory and configured to execute a computer program stored in the memory, such that the communication device performs a method according to any one of the possible implementations of the first to fourth aspects.

[0079] In a possible design solution, the communication device according to the eighth aspect may further include a transceiver. The transceiver may be a transceiver circuit or an interface circuit. The transceiver may be used by the communication device according to the eighth aspect to communicate with other communication devices.

[0080] In an embodiment of this application, the communication device according to the eighth aspect may be a device according to any one of the first to fourth aspects, or a chip (system) or other part or component that may be disposed within the device, or an apparatus including the device.

[0081] Furthermore, for the technical effects of the communication device according to the eighth aspect, refer to the technical effects of the method according to any one of the implementation methods of the first to fourth aspects, and the details will not be described again in this specification.

[0082] According to a ninth aspect, there is provided a communication device, including a processor and a memory. The memory is configured to store a computer program. When the processor executes the computer program, the communication device can execute a method according to any one of the implementation methods of the first to fourth aspects.

[0083] In a possible design solution, the communication device according to the 98th aspect may further include a transceiver. The transceiver may be a transceiver circuit or an interface circuit. The transceiver may be used by the communication device according to the 9th aspect to communicate with other communication devices.

[0084] In an embodiment of this application, the communication device according to the ninth aspect may be a device according to any one of the first to fourth aspects, or a chip (system) or other part or component that may be disposed within the device, or an apparatus including the device.

[0085] Furthermore, for the technical effects of the communication device according to the ninth aspect, refer to the technical effects of the method according to any one of the implementation methods of the first to fourth aspects, and the details will not be described again in this specification.

[0086] According to a tenth aspect, there is provided a communication system including a terminal and a network device, wherein the terminal is configured to perform the method according to the first aspect and the network device is configured to perform the method according to the second aspect, or the terminal is configured to perform the method according to the third aspect and the network device is configured to perform the method according to the fourth aspect.

[0087] According to an eleventh aspect, there is provided a computer-readable storage medium, comprising a computer program or instructions which, when executed on a computer, enable the computer to perform a method according to any one of the possible implementations of the first to fourth aspects.

[0088] According to a twelfth aspect, there is provided a computer program product, comprising a computer program or instructions which, when executed on a computer, enable the computer to carry out a method according to any one of the possible implementations of the first to fourth aspects. [Brief explanation of the drawings]

[0089] [Figure 1] FIG. [Figure 2] FIG. 1 is a diagram of an application scenario of hierarchical sweeping. [Figure 3] FIG. 1 is a diagram of a neuron. [Figure 4] A diagram of the structure of a DNN. [Figure 5] 1 is a diagram of an application scenario of beam management. [Figure 6] 1 is a diagram of an architecture of a communication system according to an embodiment of the present application; [Figure 7] 2 is an architecture of a communication system according to an embodiment of the present application. [Figure 8] 1 is a schematic flowchart 1 of a communication method according to an embodiment of the present application; [Figure 9]2 is a schematic flowchart 2 of a communication method according to an embodiment of the present application. [Figure 10] 1 is a diagram of the structure of a communication device according to an embodiment of the present application; [Figure 11] 2 is a structure of a communication device according to an embodiment of the present application; DETAILED DESCRIPTION OF THE INVENTION

[0090] In order to facilitate understanding of the technical solutions provided in the embodiments of this application, a brief description of the technology related to this application will be given first. The brief description is as follows:

[0091] 1.Beam:

[0092] With the development of fifth-generation (5G) mobile communication systems, the spectrum used is gradually evolving to higher frequency bands. Due to physical transmission characteristics, the free-space transmission loss and penetration loss in high-frequency bands are significantly higher than those in low-frequency bands. To compensate for this drawback, the transmitter or receiver of a network device or terminal concentrates energy in a narrow range through an antenna array to form a special directional transmission or reception effect, i.e., a beam, similar to the light beam formed by focusing light in a certain direction using a flashlight. Transmitting and receiving signals in the form of a beam can effectively withstand signal attenuation caused by loss and effectively increase the signal transmission distance, thereby effectively improving network coverage and increasing user experience.

[0093] The beam may be a wide beam, a narrow beam, or another type of beam. As shown in FIG. 1, a wide beam means that limited transmission energy is focused in a wide direction to balance the beam's transmission distance and coverage width. A narrow beam means that limited transmission energy is focused in a narrow direction to further increase the beam's transmission distance. The technology for forming the beam may be beamforming technology or other technology. Specifically, the beamforming technology may be digital beamforming technology, analog beamforming technology, hybrid digital / analog beamforming technology, etc.

[0094] A beam generally corresponds to a resource. For example, during beam measurement, a network device measures different beams by using different resources, and a terminal feeds back the measured resource quality, so that the network device can know the quality of the corresponding beam. During data transmission, a beam may also be indicated by using a resource corresponding to the beam. For example, a network device indicates a transmission configuration indicator state by using a transmission configuration index (TCI) field in downlink control information (DCI), and the terminal determines a beam corresponding to the reference resource based on the reference resource included in the TCI state.

[0095] In communication protocols, a beam may be specifically referred to as a digital beam, an analog beam, a spatial domain filter, a spatial filter, a spatial parameter, a TCI, a TCI state, etc. A beam used to transmit a signal may be called a transmission beam (Tx beam), a spatial domain transmission filter, a spatial transmission filter, a spatial domain transmission parameter, a spatial transmission parameter, etc. A beam used to receive a signal may be called a reception beam (Rx beam), a spatial domain reception filter, a spatial reception filter, a spatial domain reception parameter, a spatial reception parameter, etc.

[0096] Although the embodiments of this application are uniformly described using beams, it can be understood that beams may be replaced with and understood as other equivalent concepts and are not limited to the above concepts.

[0097] 2. Resources:

[0098] In a communication protocol, a reference signal is configured in the form of a resource. A network device configures each reference signal for a terminal in the form of a resource. A resource is a configuration information unit, and typically includes parameters related to the reference signal, such as the time-frequency resource position of the reference signal, the number of ports, and the time domain type (periodic / semi-static / aperiodic).

[0099] The resources may be uplink signal resources or downlink signal resources. Uplink signals include, but are not limited to, sounding reference signals (SRS) and demodulation reference signals (DMRS). Downlink signals may include, but are not limited to, channel state information reference signals (CSI-RS), cell-specific reference signals (CS-RS), user equipment-specific reference signals (US-RS), demodulation reference signals (DMRS), and synchronization signal / physical broadcast channel blocks (SS / PBCH blocks). SS / PBCH blocks may be referred to as synchronization signal blocks (SSBs) for short.

[0100] The resources may be configured by using a radio resource control (RRC) message. Regarding the configuration structure, one resource is a data structure, which includes related parameters of the uplink / downlink signal corresponding to the resource, such as the type of the uplink / downlink signal, the resource granularity carrying the uplink / downlink signal, the transmission time and transmission period of the uplink / downlink signal, and the number of ports used to transmit the uplink / downlink signal. Each uplink / downlink signal resource has a unique identifier to identify the resource of the downlink signal. It may be understood that the resource identifier may also be referred to as a resource identifier. This is not limited in the embodiments of this application.

[0101] 3.Beam Management:

[0102] In a 5G mobile communication system, both a network device and a terminal need to perform transmission by using beams. In downlink transmission, a beam used by a network device is called a downlink transmission beam, and a beam used by a terminal is called a downlink reception beam. In uplink transmission, a beam used by a terminal is called an uplink transmission beam, and a beam used by a network device is called an uplink reception beam. In downlink transmission and uplink transmission, the specific beams used by the network device and the terminal may be determined through a beam management procedure.

[0103] For example, the network device has M beams and the terminal has N beams. Then, the downlink beam management process is as follows:

[0104] (1) Beam management configuration: A network device sends a configuration message to a terminal to configure parameters related to downlink beam management for the terminal, including, for example, M measurement resources (which may also be referred to as reference signal resources) and a measurement period. The measurement resources may be reference signals (RSs) used for beam measurement. The M measurement resources have a one-to-one correspondence with the M beams of the network device. The network device transmits the corresponding measurement resource by using each beam, and the terminal measures the measurement resource to determine the quality of the beam corresponding to the measurement resource. It can be understood that the beams of the network device are invisible to the terminal, and the terminal may determine the quality of each measurement resource. However, the terminal does not sense which beam the measurement resource corresponds to.

[0105] (2) Beam Measurement: The channel quality corresponding to the beam of each network device is different from the channel quality corresponding to the beam of each terminal. The terminal needs to measure the channel quality between the beam of each network device and the beam of each terminal to determine which beam is preferable to be used by the network device for transmission and which beam is preferable to be used by the terminal for reception. Specifically, in each measurement period, the network device sequentially transmits corresponding measurement resources by using M beams, and the terminal performs reception and measurement by using one of the N beams to determine the channel quality, e.g., reference signal received power (RSRP), between the beam currently used by the terminal and the M beams. In each measurement period, the terminal may sequentially perform reception and measurement by using different beams and determine the channel quality between the N beams of the terminal and the M beams of the network device over N measurement periods.

[0106] (3) Measurement Result Report: The terminal may determine the terminal's optimal beam corresponding to each measurement resource, or the terminal's optimal receiving beam or optimal receiving beam, based on the channel quality through beam measurement. The terminal may report information about the measurement resources corresponding to the terminal's optimal beam to the network device, so that the network device determines which beam is preferable for use by the network device for transmission and which beam is preferable for use by the terminal for reception. For example, when the network device performs downlink transmission by using a beam corresponding to a specific measurement resource, the terminal performs reception by using the corresponding optimal beam. The information about the measurement resources reported by the terminal may include identifiers and RSRPs of the measurement resources corresponding to up to four optimal beams of the terminal. The network device can determine the beam of the network device corresponding to the identifiers of the measurement resources reported by the terminal based on the correspondence between the M beams and the M measurement resources. For example, the terminal may determine through beam measurement that the four measurement resources with the best quality are RS#1, RS#2, RS#3, and RS#4, and the terminal's optimal beams corresponding to the four measurement resources are Beam B1, Beam B2, Beam B2, and Beam B3, respectively. In other words, of the four measurement resources received by the terminal by using beam B1, RS#4 has the best quality. Of the four measurement resources received by the terminal by using beam B2, RS#2 and RS#3 have the best quality. Of the four measurement resources received by the terminal by using beam B3, RS#1 has the best quality. The terminal may report RS#1, RSRP1 corresponding to RS#1, RS#2, RSRP2 corresponding to RS#2, RS#3, RSRP3 corresponding to RS#3, RS#4, and RSRP4 corresponding to RS#4 to the network device.

[0107] (4) Terminal receiving beam maintenance: After the terminal reports the measurement result corresponding to the terminal's optimal beam to the network device, the terminal may maintain a mapping relationship between the terminal's optimal beam and the measurement resource corresponding to the terminal's optimal beam. For example, the mapping relationship may be as shown in Table 1 below. [Table 1]

[0108] 4.Beam direction:

[0109] For downlink transmission, the beam used by the network device needs to be notified to the terminal, so that the terminal can determine which beam to use for reception. Specifically, the network device may indicate an identifier of a measurement resource to the terminal, notifying the terminal that the network device will perform downlink transmission by using a beam corresponding to the measurement resource. In this manner, the terminal may perform reception by using the terminal's optimal beam corresponding to the measurement resource. For example, the network device may notify the terminal that the network device will perform downlink transmission by using a beam corresponding to RS#2, and the terminal will perform downlink reception by using beam B2. For uplink transmission, the beam to be used by the terminal for transmission is also indicated by the network device. Specifically, the network device may indicate an identifier of the measurement resource to the terminal, indicating that the terminal should perform uplink transmission by using the terminal's optimal beam corresponding to the measurement resource. For example, the network device may indicate to the terminal that uplink transmission will be performed by using a beam corresponding to RS#2, and the terminal will perform uplink transmission by using beam B2.

[0110] It can be seen from the above related description that narrow beams increase the transmission distance and also result in large beam management overhead, and a base station needs more narrow beams to cover the entire space. During beam management, in order to select the most suitable beam for a terminal, the terminal also needs to traverse and measure a large number of candidate beams. As a result, the beam management overhead is large. Regarding this technical problem, the prior art currently proposes implementing beam management through hierarchical sweeping to reduce the beam management overhead. The implementation method will be described in detail below.

[0111] 5. Beam management based on hierarchical sweeping:

[0112] Hierarchical sweeping typically includes two stages: a first stage is coarse sweeping, and a second stage is fine sweeping. In the first stage, a base station may perform sweeping by using a wide beam to determine the direction of a specific wide beam in which a terminal is located or to align the terminal with the wide beam. However, the accuracy of the alignment direction is not high, and the quality of the established wireless communication connection is also limited. Then, in the second stage, the base station may perform sweeping in the directions covered by the wide beam in the first stage one by one to accurately determine the direction of a specific narrow beam in which the terminal is located, thereby improving the quality of the established wireless communication connection. It can be seen that in the second stage, the sweeping beam becomes narrower, but the required sweeping range is reduced, and the number of sweeps is correspondingly reduced.

[0113] For example, as shown in FIG. 2, for the first stage, the base station performs sweeping by using wide beam ta and wide beam tb to determine that the terminal is located in the direction of wide beam ta. For the second stage, assuming that all four narrow beams can cover the corresponding wide beam, the base station may perform sweeping of the range covered by wide beam ta by using narrow beams t1 to t4 to determine that the terminal is located in the direction of narrow beam t1. In other words, the base station can accurately determine the terminal's location by sweeping only six times. Compared to sweeping all eight narrow beams, this reduces the number of sweeps by only two, thereby reducing beam management overhead.

[0114] However, with the recent emergence of artificial intelligence (AI) technology, its application in beam management has achieved significant results. Compared with beam management based on hierarchical sweeping, AI technology can further reduce beam management overhead. Details are explained below.

[0115] 6. AI-based beam management:

[0116] Machine learning is an important technological method for realizing artificial intelligence, and can be divided into supervised learning, unsupervised learning, and reinforcement learning.

[0117] From the perspective of supervised learning, based on the collected sample values ​​and sample labels, a mapping relationship between the sample values ​​and the sample labels is learned by using a machine learning algorithm, and the learned mapping relationship is expressed by using a machine learning model. The process of training a machine learning model is a process of learning a mapping relationship. For example, during signal detection, a received signal containing noise is a sample, and the actual constellation point corresponding to the signal is a label. Machine learning expects to learn the mapping relationship between the samples and the labels through training, i.e., enable the machine learning model to train a signal detector. During training, model parameters are optimized by calculating the error between the model's predicted value and the actual label. Once the mapping relationship is learned, the learned mapping can be used to predict each new sample label. The mapping relationship learned through supervised learning may include linear mapping and nonlinear mapping. Learning tasks may be classified into classification tasks and regression tasks based on the type of label.

[0118] In unsupervised learning, an algorithm is used to autonomously search for internal patterns in samples based solely on collected sample values. For certain types of algorithms in unsupervised learning, the samples are used as supervised signals. In other words, the model learns the mapping relationship between the samples, which is called self-supervised learning. During training, the model parameters are optimized by calculating the error between the model's predicted values ​​and the samples. Self-supervised learning may be used for signal compression and decompression restoration. Common algorithms include autoencoders, generative adversarial networks, etc.

[0119] Unlike supervised learning, reinforcement learning is an algorithm that learns strategies to solve problems by interacting with the environment. Unlike supervised and unsupervised learning, reinforcement learning does not have clear "correct" action label data. The algorithm needs to interact with the environment to obtain a reward signal fed back by the environment and adjust its decision action to obtain a larger reward signal value. For example, in downlink power control, a reinforcement learning model adjusts each user's downlink transmission power based on the total system throughput fed back by the wireless network in the hope of obtaining a higher system throughput. The purpose of reinforcement learning is also to learn a mapping relationship between the environmental state and the optimal decision action. However, the label for the "correct action" cannot be obtained in advance. Therefore, the network cannot optimize by calculating the error between the action and the "correct action." Reinforcement learning training is achieved through repeated interactions with the environment.

[0120] A deep neural network (DNN) is a specific form of machine learning. According to the universal approximation theorem, a neural network can theoretically approximate any continuous function, and as a result, the neural network has the ability to learn any mapping. Traditional communication systems require the design of communication modules with extensive expert knowledge. However, a DNN-based deep learning communication system can automatically discover implicit pattern structures from large datasets, establish mapping relationships between data, and achieve better performance than traditional modeling methods.

[0121] The idea behind DNNs comes from the neuron structure of the brain. Each neuron performs a weighted sum operation on its input values ​​and generates an output based on the weighted sum result by using a nonlinear function. For example, as shown in Figure 3, if the neural input is x=[x0,...,x n ] and the weights corresponding to the inputs are d=[d0,...,d n], the bias of the weighted sum is b, and the nonlinear function can be of various forms, for example, a maximum function of max{0,x}. Based on this, the execution effect of the neuron is max{0,Σ i=0 n d i x i +b}. The weight of each neuron is the model parameter of the DNN. The model parameters can be optimized during the training process, so that the DNN has the ability to extract data features and represent mapping relationships. DNNs usually optimize the model parameters according to supervised or unsupervised learning strategies.

[0122] As shown in FIG. 4, a DNN typically has a multi-layer structure. Each layer of the DNN may include multiple neurons. The input layer of the DNN processes received values ​​by using neurons and then forwards the processed values ​​to the intermediate hidden layer. A DNN typically has more than one hidden layer, and the hidden layer typically directly affects the information extraction and function fitting capabilities. Increasing the number of hidden layers of the DNN or increasing the width of each layer can improve the function fitting capabilities of the DNN. The hidden layer of the DNN then processes the received values ​​by using neurons and then forwards the calculation results to the final output layer to generate the final output of the DNN.

[0123] Based on the network construction method, DNNs may be classified into feed-forward neural networks (FNNs), convolutional neural networks (CNNs) and recurrent neural networks (RNNs).

[0124] A characteristic of FNNs is that every two neurons in adjacent layers are fully connected, which usually makes FNNs require a large amount of storage space and results in high computational complexity.

[0125] CNN is a neural network specialized for processing data with a similar grid structure. For example, both time series data (discrete sampling in the time axis) and image data (discrete sampling in two dimensions) can be considered as data with a similar grid structure. CNN performs convolution operations by capturing partial information through fixed-size windows rather than using all input information at once, which significantly reduces the amount of calculation required for model parameters. Furthermore, different convolution kernel operations can be used for each window based on different types of information captured through the window (e.g., people and objects in the same image are different types of information), allowing CNN to better extract features from the input data.

[0126] RNN is a DNN that uses feedback time series information. The input of RNN includes the new input value at the current time point and the output value of RNN at the previous time point. RNN is suitable for capturing time-correlated sequence features, especially for applications such as speech recognition, channel coding and decoding.

[0127] The above-mentioned FNN, CNN, and RNN are common neural network structures, and these network structures are constructed based on neurons. In fact, other types or structures of neural networks may also exist. Details will not be described in the embodiments of this application. Furthermore, the neural network referred to in the embodiments of this application may alternatively be replaced with any other possible description, such as a neural network model, a network model, an AI model, or an AI network model. This is not limited in this specification.

[0128] After the neural network is introduced into beam management, the terminal may perform prediction based on the neural network on which beam can be used as the optimal beam, thereby further reducing beam management overhead. Specifically, in the current AI-based beam management procedure, the terminal typically needs to configure two groups of measurement resources. One group of measurement resources may be used to realize the first round of sparse beam sweeping, designated as measurement resource group #1. The other group of measurement resources may be used to realize the second round of sweeping, designated as measurement resource group #2. A narrow beam is used as an example. The distribution pattern of the sparse beam may be as shown in FIG. 5(a). The base station may transmit eight narrow beams with different directions in the horizontal direction and eight narrow beams with different directions in the vertical direction. There are a total of 64 beams. The sparse beam may be selected from the 64 beams, and the number of sparse beams is typically 1 / 4 of the total number of beams. Alternatively, in the first round of sparse beam sweeping, a narrow beam may be replaced with a wide beam. This is not particularly limited. In the first round of sparse beam sweeping, the terminal may use a beam to receive measurement resources transmitted by the base station by using a sparse beam and determine measurement results, such as RSRP. For details, refer to the related description of beam management above. Details will not be described again in this specification. As shown in (b) of FIG. 5, the terminal may input the measurement results into a neural network to obtain indexes of K beams and report the indexes of the K beams to the base station. The K beams are beams that are most likely to be optimal beams through prediction by the neural network. The value of K may be pre-configured in the neural network, for example, 3, 4, or 5. As shown in (c) of FIG. 5, in the second round of sweeping, the base station may transmit K beams by using measurement resource group #2.The terminal may use beams to receive measurement resources transmitted by the base station by using K beams, as shown in (d) of Figure 5, determine the measurement results, and finally report the measurement resources corresponding to the optimal beam to the base station.

[0129] It can be seen that the value of K in the prior art is usually pre-configured. In practice, the number of optimal beams may be smaller than K. That is, there is a difference between the predicted number and the actual number. As a result, the indices of the K beams reported by the terminal are redundant. For example, N is smaller than K and the number of optimal beams is N. In this case, the indices of the N K beams reported by the terminal are redundant. As a result, the reporting overhead and air interface overhead are redundant, and communication efficiency is affected.

[0130] The technical solutions of this application are described below with reference to the accompanying drawings.

[0131] The technical solutions in the embodiments of this application may be applied to various communication systems, such as wireless network (Wi-Fi) systems, vehicle-to-everything (V2X) communication systems, device-to-device (D2D) communication systems, vehicular internet communication systems, 4th generation (4G) mobile communication systems, such as long term evolution (LTE) systems, worldwide interoperability for microwave access (WiMAX) communication systems, 5th generation (5G) systems, such as new radio (NR) systems, and future communication systems.

[0132] All aspects, embodiments or features are presented in this application by describing systems that may include multiple devices, components, modules, etc. It is to be appreciated and understood that each system may include other devices, components, modules, etc. and / or may not include all of the devices, components, modules, etc. discussed with reference to the accompanying drawings. Furthermore, combinations of these solutions may be used.

[0133] Furthermore, in the embodiments of this application, terms such as "example" or "for example" are used to provide an example, illustration, or explanation. Any embodiment or design manner described in this application as an "example" should not be described as preferred or having more advantages over other embodiments or design manners. Rather, the term "example" is used to present concepts in a concrete manner.

[0134] In the embodiments of this application, "information", "signal", "message", "channel" and "signaling" may be used interchangeably in some cases. It should be noted that when the difference between the terms is not emphasized, the meanings represented by the terms are equivalent. The terms "of", "corresponding, relevant", and "corresponding" may be used interchangeably in some cases. It should be noted that when the difference between the terms is not emphasized, the meanings represented by the terms are equivalent. Furthermore, " / " mentioned in this application may indicate an "or" relationship.

[0135] The network architectures and service scenarios described in the embodiments of this application are intended to more clearly explain the technical solutions in the embodiments of this application, and do not constitute limitations on the technical solutions provided in the embodiments of this application. Those skilled in the art may recognize that: with the evolution of network architectures and the emergence of new service scenarios, the technical solutions provided in the embodiments of this application can also be applied to similar technical problems.

[0136] In order to facilitate understanding of the embodiments of this application, the communication system shown in Figure 6 is first used as an example to describe in detail the communication system applicable to the embodiments of this application. For example, Figure 6 is an architecture diagram of a communication system applicable to the communication method according to the embodiments of this application.

[0137] As shown in FIG. 6, the communication system includes a terminal and a network device.

[0138] A terminal may be a terminal having receiving and transmitting capabilities, or may be a chip or chip system that may be located in a terminal. A terminal may also be called user equipment (UE), access terminal, subscriber unit, subscriber station, mobile station (MS), mobile console, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent, or user equipment. The terminal in the embodiments of this application may be a mobile phone, a cellular phone, a smartphone, a tablet computer (Pad), a wireless data card, a personal digital assistant (PDA) computer, a wireless modem, a handheld device (handset), a laptop computer, a machine type communication (MTC) terminal, a computer with wireless transmission and reception capabilities, a virtual reality (VR) terminal, an augmented reality (AR) terminal, a wireless terminal in industrial control, a wireless terminal in self-driving, a wireless terminal in remote medical, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, a wireless terminal in a smart home, an in-vehicle terminal, a road side unit (RSU) with terminal capabilities, or the like. Alternatively, the terminal in this application may be an on-board module, on-board assembly, on-board part, on-board chip, or on-board unit that is incorporated into a vehicle as one or more components or units. Alternatively, the terminal may be customer-premises equipment (CPE).

[0139] The network device may be an access network (AN) device or a radio access network (RAN) device. The RAN device may provide access functions to terminals and perform functions such as radio resource management, quality of service (QoS) management, and data compression and encryption on the air interface side. The RAN device may include a gNB in ​​5G, for example, an NR system, or one antenna panel or a group of antenna panels (including multiple antenna panels) of a base station in 5G. Alternatively, the RAN device may be a gNB, a network node forming a transmission and reception point (TRP or TP) or a transmission measurement function (TMF), for example, a baseband unit (BBU), a central unit (CU) or a distributed unit (DU), an RSU with base station functionality, a wired access gateway, or a 5G core network element. Alternatively, the RAN device may include an access point (AP) in a wireless fidelity (Wi-Fi) system, a wireless relay node, a wireless backhaul node, various types of macro base stations, micro base stations (also called small cells), relay stations, access points, wearable devices, in-vehicle devices, etc. Alternatively, the RAN device may include a next-generation mobile communication system, e.g., a 6G access network device such as a 6G base station. Alternatively, in a next-generation mobile communication system, network devices may be named in other ways, which fall within the scope of protection of the embodiments of this application, and this is not limited in this application.

[0140] It may be understood that Fig. 6 is a simplified diagram of an example for ease of understanding. The communication system may further include other network devices and / or other terminals not shown in Fig. 6.

[0141] FIG. 7 is an architecture diagram of a communication system to which a communication method according to an embodiment of this application can be applied. As shown in FIG. 7, communication between a network device and a terminal in the communication system may alternatively be represented in other forms. The terminal 10 includes a processor 101, a memory 102, and a transceiver 103. The transceiver 103 includes a transmitter 1031, a receiver 2032, and multiple antennas 1033 (antenna panels). The network device 20 includes a processor 201, a memory 202, and a transceiver 203. The transceiver 203 includes a transmitter 2031, a receiver 2032, and at least one antenna 2033 (antenna panel). The transmitter 2031 may be configured to transmit transmission control information, for example, instruction information indicating frequency division and simultaneous transmission of multiple antenna panels, to the terminal 10 through the antenna 2033. Correspondingly, the receiver 1032 may be configured to receive the transmission control information through the antenna 1033. In this manner, the transmitter 1031 may be configured to transmit transmission feedback information, i.e., uplink data, to the network device 20 in a frequency division and simultaneous transmission manner through multiple antennas 1033 (multiple antenna panels). Correspondingly, the receiver 2032 may be configured to receive the transmission feedback information transmitted by the terminal 10 through the antenna 2033.

[0142] In this embodiment of the application, after the terminal obtains RS measurement results through measurements by using a first RS resource set, the terminal may perform prediction on the RS measurement results by using a neural network model to determine a candidate RS resource set belonging to a second RS resource set. In this case, the terminal may further select a target RS resource set from the candidate RS resource sets as the RS resources to be finally reported, thereby reducing redundant overhead in the measurement process and improving communication efficiency. Alternatively, after the terminal obtains RS measurement results through measurements by using the first RS resource set, the terminal may directly report the measurement results to the network device. In this way, the network device may perform prediction on the RS measurement results by using a neural network model to determine a candidate RS resource set belonging to the second RS resource set, and further select a target RS resource set from the candidate RS resource sets as the RS resources to be finally reported, thereby reducing redundant overhead in the measurement process and improving communication efficiency.

[0143] It should be understood that further selecting a target RS resource set from the candidate RS resource sets may be performed by a terminal or may be performed by a network device. For ease of understanding, the following describes the embodiments of the method separately.

[0144] For example, Figure 8 is a schematic flowchart 1 of a communication method according to an embodiment of this application. The method is applicable to communication between a network device and a terminal in the above-mentioned communication system. In the method shown in Figure 8, the terminal may further select a target RS resource set from the candidate RS resource sets. Details will be described below.

[0145] As shown in FIG. 8, the procedure of the communication method is as follows:

[0146] S801: A terminal measures an RS from a network device by using a first RS resource set to obtain a measurement result of the RS.

[0147] The first RS resource set may be used to transmit an RS. In other words, the network device transmits an RS to the terminal by carrying the RS in the first RS resource set. Correspondingly, the terminal receives the RS carried in the first RS resource set. Transmitting an RS by the network device may also be understood as transmitting a corresponding beam. For example, the RS may be an SSB and the beam may be a wide beam. In another example, the RS may be a CSI-RS and the beam may be a narrow beam. These beams may be spatially sparse and understood as sparse beams, specifically, beams in some of the beams in all pointing directions. The sparse beams may be distributed evenly in space, unevenly, or randomly. This is not particularly limited in this specification. For example, as shown in (a) of FIG. 5, the network device may transmit eight narrow beams with different directions in the horizontal direction and eight narrow beams with different directions in the vertical direction. There are a total of 64 beams. The sparse beams may be some of the 64 beams, for example 16 beams.

[0148] The first RS resource set may be preconfigured and aligned by the network device and the terminal, e.g., preconfigured by the network device for the terminal, or preconfigured locally in the terminal and the network device according to a protocol. For the network device, the first RS resource set typically corresponds to a beam. In other words, after the network device knows the first RS resource set, the network device may know which beam or spatial direction the time-frequency resource needs to transmit the RS on. Similarly, for the terminal, the first RS resource set typically corresponds to a beam. In other words, after the terminal knows the first RS resource set, the terminal may know the time-frequency resource on which the beam or RS from the network device is received. In this way, the terminal may receive these RSs on the corresponding time-frequency resource by using each beam of the terminal, perform measurements, and obtain measurement results, e.g., RSRPs, corresponding to each beam of the terminal. The measurement results may also be referred to as measurement results of these RSs. For example, the terminal receives RSs #1 to #3 by using beam B1 and also receives RSs #1 to #3 by using beam B2. Thus, the RS measurement results may include RSRP1 when RS#1 is received using beam B1, RSRP2 when RS#2 is received using beam B1, RSRP3 when RS#3 is received using beam B1, RSRP4 when RS#1 is received using beam B2, RSRP5 when RS#2 is received using beam B2, and RSRP6 when RS#3 is received using beam B2.

[0149] S802: The terminal inputs the measurement results of the RSs into a neural network model to determine a target RS resource set.

[0150] The neural network model may be the above DNN model. For details, please refer to the related description of DNN. Alternatively, the neural network model may be any possible model, which is not limited in this specification. The neural network model may also be a trained model. For the training process of the neural network model, please refer to the related description below for understanding. The details will not be described again in this specification.

[0151] The target RS resource set may be a set of optimal RS resources. Since RS resources usually correspond to beams, the beam corresponding to the optimal RS resources is the optimal beam. For example, the beam used by a network device to transmit RS resources is the optimal transmission beam of the network device, and the beam used by a terminal to receive the optimal RS resources is the optimal receiving beam of the terminal. The optimal transmission beam of the network device and the optimal receiving beam of the terminal are a beam pair and may be used for subsequent air interface transmission between the network device and the terminal to ensure air interface transmission quality. For example, the target RS resource set includes RS resource #1 and RS resource #2. The optimal transmission beam #B1 of the network device corresponding to RS resource #1 and the optimal receiving beam #B1 of the terminal corresponding to RS resource #1 are a beam pair. The optimal transmission beam #B2 of the network device corresponding to RS resource #2 and the optimal receiving beam #B1 of the terminal corresponding to RS resource #2 are another beam pair. In this case, air interface transmission between the network device and the terminal may be achieved by using an optimal transmit beam #B1 and an optimal receive beam #B1, or by using an optimal transmit beam #B2 and an optimal receive beam #B1.

[0152] The target RS resource set may be an RS resource set determined from the candidate RS resource sets.

[0153] The candidate RS resource set may be determined by a neural network model based on RS measurement results. The candidate RS resource set may belong to a second RS resource set. The second RS resource set may also be used to transmit the RS. In other words, the network device transmits the RS to the terminal by carrying the RS in the second RS resource set. Correspondingly, the terminal receives the RS carried in the second RS resource set. In this case, the RS transmitted using the second RS resource set may be the same as or different from the RS transmitted using the first RS resource set. For example, the RS transmitted using the second RS resource set may be a CSI-RS, or the RS transmitted using the second RS resource set may be a CSI-RS or an SSB. It may be understood that all RSs transmitted using the second RS resource set are CSI-RS, and transmitting these RSs by the network device may also be understood as transmitting corresponding beams. Therefore, the second RS resource set may be the RS resource set used to transmit all narrow beams. For example, as shown in (a) of Figure 5, a network device may transmit eight narrow beams with different directions in the horizontal direction and eight narrow beams with different directions in the vertical direction. There are 64 beams in total. The second RS resource set is the RS resource set used to transmit the 64 beams.

[0154] Furthermore, similar to the first resource set, the second RS resource set may also be preconfigured and aligned by the network device and the terminal, e.g., preconfigured by the network device for the terminal, or preconfigured locally in the terminal and the network device according to a protocol. In this way, after the network device knows the second RS resource set, the network device may know which beam or spatial direction the time-frequency resource needs to be transmitted on the RS. Similarly, after the terminal knows the second RS resource set, the terminal may know the time-frequency resource on which beam or RS from the network device is received.

[0155] In this embodiment of the present application, the candidate RS resource set may be an RS resource set determined by a neural network model based on RS measurement results, specifically as follows: The information about the candidate RS resource set is an RS resource set determined by a neural network model based on RS measurement results. In other words, the terminal may input RS measurement results into the neural network model to obtain information about the candidate RS resource set output by the neural network model, i.e., information about each RS resource in the candidate RS resource set. The input to the neural network model may be all RS measurement results, for example, RSRP1 to RSRP6, or a portion of the RS measurement results, for example, RSRP1 to RSRP3. Furthermore, the number of RS resources in the candidate RS resource set may be pre-configured in the neural network model, for example, 3 to 5 or any other possible number. In other words, the neural network model may fixedly output information about 3 to 5 RS resources by calculating RS measurement results.

[0156] The information about the candidate RS resource set may be used to indicate that an RS resource in the candidate RS resource set can be used as an optimal RS, or the likelihood that an RS resource in the candidate RS resource set will be used as an optimal RS resource. In other words, possible optimal RS resources may be predicted by using a neural network model to reduce beam management overhead. For example, the information about the candidate RS resource set may include at least one of the following: the probability that each RS resource in the candidate RS resource set is an optimal RS resource; the signal quality of each RS resource in the candidate RS resource set, such as the RSRP or the RSRP of the RS carried by the RS resource; or the angle of each RS resource in the candidate RS resource set. The angle of the RS resource may be the angle difference between the transmit beam (the beam of the network device) and the receive beam (the beam of the terminal) corresponding to the RS resource, specifically, the angle difference between the transmit beam and the receive beam relative to the same reference direction. The reference direction may not be limited. For example, the direction of the transmit beam may be used as the reference method, the direction of the receive beam may be used as the reference method, or any possible direction may be used as the reference direction. It can be appreciated that a smaller angular difference indicates a higher degree of overlap between the transmit and receive beams, indicating better transmission quality.

[0157] It can be seen from the above description that when the information about the candidate RS resource sets includes the probability of an optimal RS resource, the probability of the optimal RS resource may explicitly indicate the possibility that the RS resource can be used as the optimal RS resource. In this way, the network device can perform effective resource scheduling or configuration based on the probability to avoid redundant overhead. Alternatively, when the information about the candidate RS resource sets includes the signal quality of the RS resource, the terminal needs to report the signal quality of the RS resource, e.g., the RSRP, to the network device. Thus, the signal quality of the RS resource is reused to implicitly indicate that the RS resource can be used as the optimal RS resource, thereby reducing reporting overhead. Alternatively, when the information about the candidate RS resource sets includes the angle of the RS resource, the angle of the RS resource may also explicitly indicate the possibility that the RS resource will be used as the optimal RS resource. For example, a smaller angle indicates a higher possibility that the RS resource will be used as the optimal RS resource. In this way, the network device can also perform effective resource scheduling or configuration based on the angle to avoid redundant overhead.

[0158] It can be understood that the type of information output by the neural network model depends on the training method of the model. For example, if the neural network model uses a classification training method, the output of the neural network model is a classification or a probability of a specific type, i.e., in this embodiment of the present application, the probability that the RS resource is an optimal RS resource. In another example, if the neural network model uses a regression training method, the output of the neural network model is a regression or a specific value, i.e., in this embodiment of the present application, the signal quality of the RS resource or the angle of the RS resource.

[0159] It can be further understood that the above describes the inference process of the trained neural network model in the actual application process. For ease of understanding, the training process of the neural network model is described in this specification based on the principle of the inference process. In fact, the training process is almost the same as the inference process, except that the training process further includes a backpropagation step, and the training process realizes the training of the neural network model through backpropagation.

[0160] For example, taking a classification training method as an example, the information output by the neural network model is the probability that the RS resource is the optimal RS resource, e.g., 0.7. In the training sample, the RS resource is actually the optimal RS resource. That is, the actual probability is 1. In this case, the difference between the information output by the neural network model and the training sample is 1-0.7=0.3. Therefore, the difference can be backpropagated to the neural network model to adjust the neuron weights in the neural network model, thereby achieving one training session. Through successive iterations of the training process, the output of the neural network model for an RS resource that is actually the optimal RS resource gradually approaches 1, and the output of the neural network model for an RS resource that is not actually the optimal RS resource gradually approaches 0. When the accuracy of the output of the neural network model reaches a threshold, e.g., greater than 99.99%, the neural network model is considered to have converged and the training is complete.

[0161] In another example, taking a regression training method as an example, the information output by the neural network model is the signal quality of the RS resource, for example, 3 dB. In the training sample, the actual signal quality of the RS resource is 5 dB. In this case, the difference between the information output by the neural network model and the training sample is 5 dB - 3 dB = 2 dB. Therefore, the difference can be backpropagated to the neural network model to adjust the neuron weights in the neural network model, thereby achieving one training. Through successive iterations of the training process, the signal quality output by the neural network model for the RS resource gradually approaches the actual signal quality of the RS resource. When the signal quality output by the neural network model for the RS resource is slightly different from or even the same as the actual signal quality of the RS resource, the neural network model is considered to achieve an accurate output. When the accuracy of the output of the neural network model reaches a threshold, for example, greater than 99.99%, the neural network model is considered to have converged and training is completed.

[0162] In this embodiment of the present application, the target RS resource set may specifically be an RS resource set determined from the candidate RS resource sets based on the first threshold and information about the candidate RS resource sets.

[0163] The first threshold may be related to information about the candidate RS resource set. For example, the first threshold may be set based on the information type of the candidate RS resource set (see the related description below for details), so that redundant RSs in the candidate RS resource set can be more accurately screened out. For example, the information about the candidate RS resource set is the probability that the RS resource is an optimal RS resource, and the probability represented by the first threshold may be the minimum probability that the RS resource is used as the optimal RS resource. In another example, the information about the candidate RS resource set is the signal quality of the RS resource, and the signal quality represented by the first threshold may be the minimum signal quality of the RS resource used as the optimal RS resource. In another example, the information about the candidate RS resource set is the angle of the RS resource, and the angle represented by the first threshold may be the minimum angle of the RS resource used as the optimal RS resource.

[0164] The first threshold may be a pre-configured threshold. For example, the first threshold may be pre-configured for the terminal by the network side and locally stored in the terminal. Specifically, the terminal may receive instruction information from a network device, and the instruction information may indicate the first threshold and realize flexible configuration according to actual requirements. Alternatively, the first threshold may be pre-configured (stored) locally in the terminal or pre-configured in a neural network model to avoid unnecessary overhead caused by the network side separately configuring the first threshold.

[0165] When the first threshold is stored locally in the terminal, the terminal may select information about the candidate RS resource sets based on the first threshold. Alternatively, when the first threshold is configured in a neural network model, the neural network model may select information about the candidate reference signal resource sets based on the first threshold. In this case, the information about the determined target RS resource set and the first threshold satisfy at least one of the following relationships: the probability that an RS resource in the target RS resource set is an optimal RS resource is greater than the probability represented by the first threshold; the signal quality of the RS resource in the target RS resource set is greater than the signal quality represented by the first threshold; or the angle of each RS resource in the RS resource set is less than the angle represented by the first threshold. In other words, the RS resource most likely to be used as the optimal RS resource is selected from the candidate RS resource sets based on the first threshold, thereby avoiding redundant overhead.

[0166] It can be understood that the target RS resource set is a candidate RS resource set if the information about all RS resources in the candidate RS resource set satisfies the first threshold. For example, the candidate RS resource set includes RS resource #1 to RS resource #4. The RSRP of RS resource #1 is 4 dB, the RSRP of RS resource #2 is 5 dB, the RSRP of RS resource #3 is 4.2 dB, and the RSRP of RS resource #4 is 4.6 dB. If the RSRP represented by the first threshold is 3.5 dB, the RSRPs of RS resource #1 to RS resource #4 are all greater than the RSRP represented by the first threshold. Therefore, RS resource #1 to RS resource #4 may be selected. That is, the target RS resource set also includes RS resource #1 to RS resource #4.

[0167] Alternatively, if information about only some RS resources in the candidate RS resource set satisfies the first threshold, the target RS resource set includes some RS resources that satisfy the first threshold and are in the candidate RS resource set. For example, the candidate RS resource set includes RS resource #1 to RS resource #4. The RSRP of RS resource #1 is 4 dB, the RSRP of RS resource #2 is 5 dB, the RSRP of RS resource #3 is 4.2 dB, and the RSRP of RS resource #4 is 4.6 dB. If the RSRP represented by the first threshold is 4.5 dB, only the RSRPs of RS resource #2 and RS resource #4 are greater than the RSRP represented by the first threshold. Therefore, RS resource #2 and RS resource #4 may be selected. That is, the target RS resource set includes RS resource #2 and RS resource #4 but does not include RS resource #1 and RS resource #3.

[0168] Alternatively, if the information about any RS resource in the candidate RS resource set does not satisfy the first threshold, this indicates that the value of the first threshold may be inappropriate, and the terminal needs to reconfigure the first threshold with a more appropriate value.

[0169] In conclusion, after the terminal obtains the measurement results of the RSs through measurements by using the first RS resource set, the terminal may perform prediction on the measurement results of the RSs by using a neural network model to determine a candidate RS resource set belonging to the second RS resource set, in which case the terminal may further select a target RS resource set from the candidate RS resource sets as the RS resources to be finally reported, thereby reducing redundant overhead in the measurement process and improving communication efficiency.

[0170] Optionally, referring to S801 and S802, the communication method provided in this embodiment of this application may further include the following steps:

[0171] S803: The terminal sends information about the target RS resource set to the network device.

[0172] The information about the target RS resource set transmitted by the terminal to the network device is used by the network device to determine which RS resources can be used as optimal RS resources. The information about the target RS resource set transmitted by the terminal to the network device may include at least one of the following: identifiers of RS resources in the target RS resource set, signal qualities of RS resources in the target RS resource set, or angles of each RS resource in the RS resource set. For example, in some cases, the information about the target RS resource set may include identifiers and signal qualities of RS resources, which together indicate which RS resources can be used as optimal RS resources. Alternatively, in other cases, the information about the target RS resource set may include angles of RS resources. In this way, the network device may also determine which RS resources can be used as optimal RS resources based only on the angles of the RS resources.

[0173] It may be understood that the network device may determine which RS resource can be used as the optimal RS resource based on the information about the target RS resource set in a conventional manner. For details, please refer to the related description in "6. AI-Based Beam Management." The details will not be described again in this specification.

[0174] Optionally, in the communication method provided in this embodiment of the present application, there may be multiple target RS resource sets, denoted as target RS resource set groups, and the multiple target RS resource sets in the target RS resource set group may be configured for use separately in different time units to ensure that optimal RS resources are available for air interface transmission in each time unit, thereby ensuring the stability and reliability of air interface transmission. In this case, the communication method provided in this embodiment of the present application may be replaced by being realized by performing steps A to C. Details will be described below.

[0175] Step A: The terminal measures the RS from the network device by using a first reference signal RS resource set to obtain a measurement result of the RS.

[0176] The terminal may separately perform measurements in multiple time units to obtain measurement results of the RS in each time unit. The multiple time units may be adjacent time units or may not be adjacent time units. This is not limited in this specification. Furthermore, the specific principle of the terminal obtaining measurement results of the RS in each time unit through measurements in each time unit is the same as the principle in S801. For details, please refer to the above description. The details will not be described again in this specification.

[0177] It can be understood that the time unit is merely an exemplary name for ease of explanation in this embodiment, and may alternatively be replaced with any possible name, such as a time window, a sliding time window, or a period, which is not particularly limited.

[0178] Step B: The terminal inputs the measurement result into a neural network model to determine a target RS resource set group.

[0179] The target RS resource set group may include target RS resource sets corresponding to multiple time units, respectively, or may include multiple target RS resource sets separately configured for use in different time units, such as a first target RS resource set corresponding to a first time unit and a second target RS resource set corresponding to a second time unit. The first time unit and the second time unit are different time units, such as two adjacent time units. The first target RS resource set is determined from a first candidate RS resource set, and the second target RS resource set is determined from a second candidate RS resource set. The first candidate RS resource set and the second candidate RS resource set are determined by a neural network model based on measurement results, and both the first candidate RS resource set and the second candidate RS resource set belong to the second RS resource set. Details are described below.

[0180] Specifically, the terminal may input some or all of the measurement results determined in step A into a neural network module to obtain information about the candidate RS resource set group, for example, including information about the first candidate RS resource set and information about the second candidate RS resource set. In other words, the information about the first candidate RS resource set is determined by the neural network model based on the measurement results of the RSs in step A, and the information about the second candidate RS resource set is determined by the neural network model based on the measurement results of the RSs in step A.

[0181] The information about the first candidate RS resource set may include at least one of the following: a probability that each RS resource in the first candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the first candidate RS resource set, or an angle of each RS resource in the first candidate RS resource set, which may be an angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0182] The information about the second candidate RS resource set may include at least one of the following: a probability that each RS resource in the first candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the first candidate RS resource set, or an angle of each RS resource in the first candidate RS resource set, which may be the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0183] The terminal or the neural network model may obtain a first target RS resource set by filtering the first candidate RS resource set based on the first threshold and information about the first candidate RS resource set. In other words, the first target RS resource set is determined from the first candidate RS resource set based on the first threshold and information about the first candidate RS resource set. The first threshold is related to the information about the first candidate RS resource set. In this case, the information about the first target RS resource set and the first threshold may satisfy at least one of the following relationships: a probability that an RS resource in the first target RS resource set is an optimal RS resource is greater than a probability represented by the first threshold; a signal quality of the RS resource in the first target RS resource set is greater than a signal quality represented by the first threshold; or an angle of each RS resource in the first RS resource set is less than an angle represented by the first threshold.

[0184] Similarly, the terminal or the neural network model may also filter the second candidate RS resource set based on the second threshold and information about the second candidate RS resource set to obtain a second target RS resource set. In other words, the second target RS resource set is determined from the second candidate RS resource set based on the second threshold and information about the second candidate RS resource set. The second threshold is related to the information about the second candidate RS resource set. In this case, the information about the second target RS resource set and the second threshold may also satisfy at least one of the following relationships: the probability that the RS resources in the first target RS resource set are optimal RS resources is greater than the probability represented by the first threshold, the signal quality of the RS resources in the first target RS resource set is greater than the signal quality represented by the first threshold, or the angle of each RS resource in the first RS resource set is less than the angle represented by the first threshold.

[0185] It may be understood that the first threshold and the second threshold may be the same threshold. In this case, for configuration manners of the first threshold or the second threshold, refer to the related description in 902. Alternatively, the first threshold and the second threshold may be different thresholds. In this case, the first threshold and the second threshold may be configured separately. For example, the terminal may receive first instruction information from the network device. The first instruction information may indicate the first threshold, or the first threshold may be preconfigured in the neural network model, or may be preconfigured locally in the terminal. In another example, the terminal may also receive second instruction information from the network device. The second instruction information may indicate the second threshold, or the second threshold may be preconfigured in the neural network model, or may be preconfigured locally in the terminal.

[0186] It can be further understood that the determination of each target RS resource set in step B is similar to the determination in S802. For details, please refer to the above description. The details will not be described again in this specification.

[0187] It may be further understood that step A and step B may be performed in a polling manner. For example, the terminal separately performs measurements at time unit t-2 and time unit t-1, and inputs the RS measurement results into the neural network model to obtain target RS resource sets corresponding to time unit t and time unit t+1, respectively. Then, the terminal separately performs measurements at time unit t and time unit t+1, and inputs the RS measurement results into the neural network model to obtain target RS resource sets corresponding to time unit t+2 and time unit t+3, respectively, and so on.

[0188] (Optionally) Step C: The terminal sends information about the target RS resource set group to the network device.

[0189] For example, the target RS resource set group includes a first target RS resource set and a second target RS resource set. The information about the first target RS resource set sent by the terminal to the network device may include at least one of the following: identifiers of RS resources in the first target RS resource set, signal qualities of RS resources in the first target RS resource set, or angles of each RS resource in the first target RS resource set. For specific implementation manners, please refer to the related description in 903. Details will not be described again in this specification. The information about the second target RS resource set sent by the terminal to the network device may include at least one of the following: identifiers of RS resources in the second target RS resource set, signal qualities of RS resources in the second target RS resource set, or angles of each RS resource in the second target RS resource set. For specific implementation manners, please refer to the related description in 903. Details will not be described again in this specification.

[0190] For example, Figure 9 is a schematic flowchart 2 of a communication method according to an embodiment of this application. The method is applicable to communication between a network device and a terminal in the above communication system. In the method shown in Figure 9, the network device may further select a target RS resource set from the candidate RS resource sets. Details will be described below.

[0191] As shown in FIG. 9, the procedure of the communication method is as follows:

[0192] S901: A network device transmits an RS to a terminal by using a first reference signal RS resource set.

[0193] S902: The network device receives a measurement result of the RS fed back by the terminal.

[0194] The measurement results of the RSs fed back by the terminal may be all the measurement results or a part of the measurement results. In other words, the terminal may choose to report only a part of the measurement results to reduce overhead. Furthermore, the specific implementation manners of S901 and S902 are the same as the implementation manner of 901. The difference is that the terminal still needs to feed back the measurement results to the network device. For other details, please refer to the above description. The details will not be described again in this specification.

[0195] S903: The network device inputs the measurement result into a neural network model to determine a target RS resource set.

[0196] The target RS resource set is determined from the candidate RS resource sets. The candidate RS resource set is determined by a neural network model based on measurement results. The candidate RS resource set belongs to a second RS resource set. The RS transmitted by using the first RS resource set is the same as or different from the RS transmitted by using the second RS resource set.

[0197] In a possible design solution, the target RS resource set may be determined from the candidate RS resource sets based on a first threshold and information about the candidate RS resource sets, where the first threshold is related to the information about the candidate RS resource sets.

[0198] Optionally, the candidate RS resource set is determined by a neural network model based on the measurement results of RSs, specifically as follows: Information about the candidate RS resource set is determined by a neural network model based on the measurement results.

[0199] Furthermore, the information about the candidate RS resource sets includes at least one of the following: a probability that each RS resource in the candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the candidate RS resource set, or an angle of each RS resource in the candidate RS resource set, where the angle of each RS resource in the candidate RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0200] Furthermore, the information about the target RS resource set and the first threshold satisfy at least one of the following relationships: the probability that the RS resources in the target RS resource set are optimal RS resources is greater than the probability represented by the first threshold, the signal quality of the RS resources in the target RS resource set is greater than the signal quality represented by the first threshold, or the angle of each RS resource in the RS resource set is less than the angle represented by the first threshold.

[0201] Optionally, the first threshold may be pre-configured in the neural network model or may be pre-configured locally in the network device.

[0202] It can be understood that the specific implementation principle of S903 is similar to that of S802. The difference is that the execution body of S903 is replaced by a network device. For other details, please refer to the above description. The details will not be described again in this specification.

[0203] In conclusion, after the terminal obtains the RS measurement results through measurements by using the first RS resource set, the terminal may directly report the measurement results to the network device. In this way, the network device may perform prediction on the RS measurement results by using a neural network model to determine candidate RS resource sets belonging to the second RS resource set, and further select a target RS resource set from the candidate RS resource sets as the RS resource to be finally reported, thereby reducing redundant overhead in the measurement process and improving communication efficiency.

[0204] Optionally, referring to S901 to S903, the communication method provided in this embodiment of this application may further include the following steps:

[0205] S904: The network device sends the target RS resource set to the terminal.

[0206] Specifically, the network device may send the target RS resource set to the terminal as follows: The network device sends an identifier of each RS resource in the target RS resource set to the terminal, instructing the network device to transmit a beam on the target RS resource set, thereby ensuring that the terminal is ready for reception and measurement. Then, the terminal further reports information about the target RS resource set to the network device based on the reception and measurement. In this way, the network device's determination of which RS resource can be used as the optimal RS resource based on the information about the target RS resource set may be realized in a manner in the prior art. For details, please refer to the related description in "6. AI-Based Beam Management." The details will not be described again in this specification.

[0207] Optionally, in the communication method provided in this embodiment of the present application, there may also be multiple target RS resource sets, denoted as target RS resource set groups, and the multiple target RS resource sets in the target RS resource set group may be configured for use separately in different time units to ensure that optimal RS resources are available for air interface transmission in each time unit, thereby ensuring the stability and reliability of air interface transmission. In this case, the communication method provided in this embodiment of the present application may be replaced by being realized by performing steps 1 to 4. Details will be described below.

[0208] Step 1: A network device transmits an RS to a terminal by using a first reference signal RS resource set.

[0209] Step 2: The network device receives the measurement result of the RS fed back by the terminal.

[0210] The terminal may separately perform measurements in multiple time units to obtain measurement results of the RS in each time unit. The multiple time units may be adjacent time units or may not be adjacent time units. This is not limited in this specification. Furthermore, the specific principle of the terminal obtaining the measurement results of the RS in each time unit through measurements in each time unit is the same as the principle in S801. For details, please refer to the above description. The details will not be described again in this specification. Furthermore, similar to S901 and S902, the measurement results of the RS fed back by the terminal may be all measurement results or a part of the measurement results.

[0211] Step 3: The network device inputs the measurement results into the neural network model to determine the target RS resource set group.

[0212] The specific implementation of step 3 is the same as that of step B. The difference is that the execution body is replaced by a network device. For other details, please refer to the above description. The details will not be described again in this specification.

[0213] It may be understood that steps 1 to 3 may be performed in a polling manner. For example, the network device separately performs measurements at time unit t-2 and time unit t-1, and inputs the RS measurement results into the neural network model to obtain target RS resource sets corresponding to time unit t and time unit t+1, respectively. Then, the network device separately performs measurements at time unit t and time unit t+1, and inputs the RS measurement results into the neural network model to obtain target RS resource sets corresponding to time unit t+2 and time unit t+3, respectively, and so on.

[0214] (Optionally) Step 4: The network device sends information about the target RS resource set group to the terminal.

[0215] For example, the target RS resource set group includes a first target RS resource set and a second target RS resource set. The information about the first target RS resource set transmitted by the network device to the terminal may include identifiers of RS resources in the first target RS resource set. The information about the second target RS resource set transmitted by the network device to the terminal may include identifiers of RS resources in the second target RS resource set.

[0216] Furthermore, the specific implementation manner of step 4 is the same as that of step C. For details, please refer to the above description. The details will not be described again in this specification.

[0217] Above, the communication method provided in the embodiment of this application is described in detail with reference to Figures 8 and 9. Hereinafter, the communication device configured to perform the communication method provided in the embodiment of this application is described in detail with reference to Figures 10 and 11.

[0218] 10 is a diagram of the structure of a communication device according to an embodiment of this application. For example, as shown in FIG. 10, the communication device 1000 includes a transceiver module 1001 and a processing module 1002. The transceiver module 1001 is configured to perform transceiver functions of the communication device 1000, and the processing module 1002 is configured to perform functions other than the transceiver functions of the communication device 1000.

[0219] For ease of explanation, FIG. 10 shows only the main components of the communication device.

[0220] In some embodiments, the communication device 1000 may be applicable to the communication systems shown in FIGS. 6 and 7 and may perform the functions of a terminal in the method shown in FIG.

[0221] for example,

[0222] The processing module 1002 is configured to control the transceiver module 1001 to measure RSs from network devices by using a first reference signal RS resource set to obtain measurement results of the RSs. The processing module 1002 is further configured to input the measurement results into a neural network model to determine a target RS resource set. The target RS resource set is determined from candidate RS resource sets, and the candidate RS resource set is determined by the neural network model based on the measurement results, and the candidate RS resource set belongs to a second RS resource set. The RSs transmitted by using the first RS resource set may be the same as or different from the RSs transmitted by using the second RS resource set.

[0223] In a possible design solution, the target RS resource set is determined from the candidate RS resource sets based on a first threshold and information about the candidate RS resource sets, where the first threshold is related to the information about the candidate RS resource sets.

[0224] Optionally, the candidate RS resource set is determined by a neural network model based on the measurement results of the RSs, specifically as follows: Information about the candidate RS resource set is determined by a neural network model based on the measurement results of the RSs.

[0225] Optionally, the information about the candidate RS resource sets may include at least one of the following: a probability that each RS resource in the candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the candidate RS resource set, or an angle of each RS resource in the candidate RS resource set, where the angle of each RS resource in the candidate RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0226] Furthermore, the information about the target RS resource set and the first threshold satisfy at least one of the following relationships: the probability that the RS resources in the target RS resource set are optimal RS resources is greater than the probability represented by the first threshold, the signal quality of the RS resources in the target RS resource set is greater than the signal quality represented by the first threshold, or the angle of each RS resource in the RS resource set is less than the angle represented by the first threshold.

[0227] Optionally, the transceiver module 1001 is further configured to receive indication information from the network device, the indication information indicating the first threshold value.

[0228] Optionally, the first threshold may alternatively be pre-configured in the neural network model.

[0229] In a possible design solution, the transceiver module 1001 is further configured to transmit information about the target RS resource set to the network device. The information about the target RS resource set includes at least one of the following: identifiers of RS resources in the target RS resource set, signal qualities of RS resources in the target RS resource set, or angles of each RS resource in the RS resource set. The angle of each RS resource in the target RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0230] In a possible design solution, there are multiple target RS resource sets, which are separately configured for use in different time units.

[0231] In other examples,

[0232] The processing module 1002 is configured to control the transceiver module 1001 to measure RSs from network devices by using a first reference signal RS resource set to obtain RS measurement results. The processing module 1002 is further configured to input the measurement results into a neural network model to determine a target RS resource set group. The target RS resource set group includes a first target RS resource set corresponding to a first time unit and a second target RS resource set corresponding to a second time unit, where the first time unit is different from the second time unit. The first target RS resource set is determined from a first candidate RS resource set, and the second target RS resource set is determined from the second candidate RS resource set. The first candidate RS resource set and the second candidate RS resource set are determined by the neural network model based on the measurement results, where both the first candidate RS resource set and the second candidate RS resource set belong to the second RS resource set. The RS transmitted using the first RS resource set may be the same as or different from the RS transmitted using the second RS resource set.

[0233] In a possible design solution, a first target RS resource set corresponding to a first time unit means that the first target RS resource set is configured for use in the first time unit.

[0234] In a possible design solution, the second target RS resource set corresponding to the second time unit means that the second target RS resource set is configured for use in the second time unit.

[0235] In a possible design solution, a first target RS resource set is determined from a first candidate RS resource set based on a first threshold and information about the first candidate RS resource set, where the first threshold is related to the information about the first candidate RS resource set.

[0236] Optionally, the first candidate RS resource set is determined by a neural network model based on measurement results of RSs, specifically as follows: Information about the first candidate RS resource set is determined by a neural network model based on measurement results of RSs.

[0237] Optionally, the information about the first candidate RS resource set includes at least one of the following: a probability that each RS resource in the first candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the first candidate RS resource set, or an angle of each RS resource in the first candidate RS resource set, where the angle of each RS resource in the first candidate RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0238] Furthermore, the information about the first target RS resource set and the first threshold satisfy at least one of the following relationships: the probability that the RS resources in the first target RS resource set are optimal RS resources is greater than the probability represented by the first threshold, the signal quality of the RS resources in the first target RS resource set is greater than the signal quality represented by the first threshold, or the angle of each RS resource in the first RS resource set is less than the angle represented by the first threshold.

[0239] Optionally, the transceiver module 1001 is further configured to receive first indication information from the network device, the first indication information indicating the first threshold value.

[0240] Optionally, the first threshold may alternatively be pre-configured in the neural network model.

[0241] In a possible design solution, the transceiver module 1001 is further configured to transmit information about the first target RS resource set to the network device. For example, the information about the first target RS resource set includes at least one of the following: identifiers of RS resources in the first target RS resource set, signal qualities of RS resources in the first target RS resource set, or angles of each RS resource in the first RS resource set. The angle of each RS resource in the first target RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0242] In a possible design solution, the second target RS resource set is determined from the second candidate RS resource set based on a second threshold and information about the second candidate RS resource set, where the second threshold is related to the information about the second candidate RS resource set.

[0243] Optionally, the second candidate RS resource set is determined by a neural network model based on the measurement results of the RSs, specifically as follows: Information about the second candidate RS resource set is determined by a neural network model based on the measurement results of the RSs.

[0244] Optionally, the information about the second candidate RS resource set includes at least one of the following: a probability that each RS resource in the second candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the second candidate RS resource set, or an angle of each RS resource in the second candidate RS resource set, where the angle of each RS resource in the candidate RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0245] Optionally, the transceiver module 1001 is further configured to receive second indication information from the network device, the second indication information indicating the second threshold value.

[0246] Optionally, the second threshold may alternatively be pre-configured in the neural network model.

[0247] In a possible design solution, the transceiver module 1001 is further configured to transmit information about the second target RS resource set to the network device. For example, the information about the second target RS resource set includes at least one of the following: an identifier of the RS resource in the second target RS resource set, a signal quality of the RS resource in the second target RS resource set, or an angle of each RS resource in the second RS resource set. The angle of each RS resource in the second target RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0248] Optionally, the transceiver module 1001 may include a transmitting module (not shown in FIG. 10) and a receiving module (not shown in FIG. 10). The transmitting module is configured to realize a transmitting function of the communication device 1000, and the receiving module is configured to realize a receiving function of the communication device 1000.

[0249] Optionally, the communication device 1000 may further include a storage module (not shown in FIG. 10) for storing programs or instructions. When the processing module 1002 executes the programs or instructions, the communication device 1000 is enabled to perform the functions of a terminal in the method in FIG.

[0250] It may be understood that the communication device 1000 may be a terminal, a chip (system) or other part or component that may be disposed in a terminal, or a device that includes a terminal, which is not limited in this application.

[0251] Furthermore, for the technical effects of the communication device 1000, please refer to the technical effects of the communication method shown in Fig. 9. The details will not be described again in this specification.

[0252] In some other embodiments, the communication device 1000 may be applicable to the communication systems shown in FIGS. 6 and 7 and may perform the functions of a network device in the method shown in FIG.

[0253] for example,

[0254] The transceiver module 1001 is configured to transmit an RS to a terminal by using a first reference signal RS resource set and receive measurement results of the RS fed back by the terminal. Thus, the processing module 1002 is configured to input the measurement results into a neural network model to determine a target RS resource set. The target RS resource set is determined from candidate RS resource sets, and the candidate RS resource set is determined by the neural network model based on the measurement results, where the candidate RS resource set belongs to a second RS resource set, and the RS transmitted by using the first RS resource set is the same as or different from the RS transmitted by using the second RS resource set.

[0255] In a possible design solution, the target RS resource set is determined from the candidate RS resource sets based on a first threshold and information about the candidate RS resource sets, where the first threshold is related to the information about the candidate RS resource sets.

[0256] Optionally, the candidate RS resource set is determined by a neural network model based on the measurement results of RSs, specifically as follows: Information about the candidate RS resource set is determined by a neural network model based on the measurement results.

[0257] Furthermore, the information about the candidate RS resource sets includes at least one of the following: a probability that each RS resource in the candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the candidate RS resource set, or an angle of each RS resource in the candidate RS resource set, where the angle of each RS resource in the candidate RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0258] Furthermore, the information about the target RS resource set and the first threshold satisfy at least one of the following relationships: the probability that the RS resources in the target RS resource set are optimal RS resources is greater than the probability represented by the first threshold, the signal quality of the RS resources in the target RS resource set is greater than the signal quality represented by the first threshold, or the angle of each RS resource in the RS resource set is less than the angle represented by the first threshold.

[0259] Optionally, the first threshold may be pre-configured in the neural network model.

[0260] In a possible design solution, the transceiver module 1001 is further configured to transmit the target RS resource set to the terminal.

[0261] In a possible design solution, there are multiple target RS resource sets, and the multiple target RS resource sets are configured for use in different time units.

[0262] In a possible design solution, the RS measurement is the entire measurement or a part of the measurement.

[0263] In other examples,

[0264] The transceiver module 1001 is configured to transmit an RS to a terminal by using a first reference signal RS resource set and receive a measurement result of the RS fed back by the terminal. The processing module 1002 is configured to input the measurement result into a neural network model to determine a target RS resource set group. The target RS resource set group includes a first target RS resource set corresponding to a first time unit and a second target RS resource set corresponding to a second time unit, where the first time unit is different from the second time unit. The first target RS resource set is determined from a first candidate RS resource set, and the second target RS resource set is determined from the second candidate RS resource set. The first candidate RS resource set and the second candidate RS resource set are determined by the neural network model based on the measurement result, where both the first candidate RS resource set and the second candidate RS resource set belong to the second RS resource set. The RS transmitted using the first RS resource set may be the same as or different from the RS transmitted using the second RS resource set.

[0265] In a possible design solution, a first target RS resource set corresponding to a first time unit means that the first target RS resource set is configured for use in the first time unit.

[0266] In a possible design solution, the second target RS resource set corresponding to the second time unit means that the second target RS resource set is configured for use in the second time unit.

[0267] In a possible design solution, a first target RS resource set is determined from a first candidate RS resource set based on a first threshold and information about the first candidate RS resource set, where the first threshold is related to the information about the first candidate RS resource set.

[0268] Optionally, the first candidate RS resource set is determined by a neural network model based on measurement results of RSs, specifically as follows: Information about the first candidate RS resource set is determined by a neural network model based on measurement results of RSs.

[0269] Furthermore, the information about the first candidate RS resource set includes at least one of the following: a probability that each RS resource in the first candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the first candidate RS resource set, or an angle of each RS resource in the first candidate RS resource set, where the angle of each RS resource in the first candidate RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0270] Furthermore, the information about the first target RS resource set and the first threshold satisfy at least one of the following relationships: the probability that the RS resources in the first target RS resource set are optimal RS resources is greater than the probability represented by the first threshold, the signal quality of the RS resources in the first target RS resource set is greater than the signal quality represented by the first threshold, or the angle of each RS resource in the first RS resource set is less than the angle represented by the first threshold.

[0271] Optionally, the first threshold may be pre-configured in the neural network model.

[0272] In a possible design solution, the transceiver module 1001 is further configured to transmit information regarding the first target RS resource set to the network device.

[0273] In a possible design solution, the second target RS resource set is determined from the second candidate RS resource set based on a second threshold and information about the second candidate RS resource set, where the second threshold is related to the information about the second candidate RS resource set.

[0274] Optionally, the second candidate RS resource set is determined by a neural network model based on the measurement results of the RSs, specifically as follows: Information about the second candidate RS resource set is determined by a neural network model based on the measurement results of the RSs.

[0275] Furthermore, the information about the second candidate RS resource set includes at least one of the following: a probability that each RS resource in the second candidate RS resource set is an optimal RS resource, a signal quality of each RS resource in the second candidate RS resource set, or an angle of each RS resource in the second candidate RS resource set, where the angle of each RS resource in the second candidate RS resource set is the angle difference between the transmit beam and the receive beam corresponding to the RS resource.

[0276] Furthermore, the information about the second target RS resource set and the second threshold satisfy at least one of the following relationships: the probability that the RS resources in the second target RS resource set are optimal RS resources is greater than the probability represented by the second threshold, the signal quality of the RS resources in the second target RS resource set is greater than the signal quality represented by the second threshold, or the angle of each RS resource in the second RS resource set is less than the angle represented by the second threshold.

[0277] Optionally, the second threshold may be pre-configured in the neural network model.

[0278] In a possible design solution, the transceiver module 1001 is further configured to transmit the second target RS resource set to the terminal.

[0279] In a possible design solution, the RS measurement is the entire measurement or a part of the measurement.

[0280] Optionally, the transceiver module 1001 may include a transmitting module (not shown in FIG. 10) and a receiving module (not shown in FIG. 10). The transmitting module is configured to realize a transmitting function of the communication device 1000, and the receiving module is configured to realize a receiving function of the communication device 1000.

[0281] Optionally, the communication device 1000 may further include a storage module (not shown in FIG. 10 ) that stores programs or instructions. When the processing module 1002 executes the programs or instructions, the communication device 1000 is enabled to perform the functions of a network device in the method in FIG. 10 .

[0282] It may be understood that the communication device 1000 may be a terminal, a chip (system) or other part or component that may be disposed in a terminal, or a device that includes a terminal, which is not limited in this application.

[0283] Furthermore, for the technical effects of the communication device 1000, please refer to the technical effects of the communication method shown in Fig. 10. The details will not be described again in this specification.

[0284] FIG. 11 is a diagram illustrating a structure of a communication device according to an embodiment of this application. For example, the communication device may be a terminal, or may be a chip (system) or other part or component that may be disposed in a terminal. As shown in FIG. 11 , the communication device 1100 may include a processor 1101. Optionally, the communication device 1100 may further include a memory 1102 and / or a transceiver 1103. The processor 1101 is coupled to the memory 1102 and the transceiver 1103, and may be connected, for example, through a communication bus. Furthermore, the communication device 1100 may also be a chip, for example, including the processor 1101. In this case, the transceiver may be an output / input interface of the chip.

[0285] The components of the communication device 1100 are described in more detail below with reference to FIG.

[0286] Processor 1101 is the control center of communication device 1100 and may be a single processor or a collective term for multiple processing elements. For example, processor 1101 may be one or more central processing units (CPUs), an application-specific integrated circuit (ASIC), or one or more integrated circuits for implementing embodiments of this application, such as one or more microprocessors (digital signal processors, DSPs) or one or more field programmable gate arrays (FPGAs).

[0287] Optionally, the processor 1101 may operate or execute software programs stored in the memory 1102 and access data stored in the memory 1102 to perform various functions of the communication device 1100, for example, to perform the communication methods described above.

[0288] In a specific implementation, in an embodiment, the processor 1101 may include one or more CPUs, for example, CPU0 and CPU1 shown in FIG.

[0289] In a specific implementation, in an embodiment, communications device 1100 may alternatively include multiple processors, such as processor 1101 and processor 1104 shown in FIG. 11. Each processor may be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). A processor herein may refer to one or more devices, circuits, and / or processing cores configured to process data (e.g., computer programs or instructions).

[0290] The memory 1102 is configured to store a software program for implementing the solution in this application, and the processor 1101 controls the execution. For specific implementation manners, please refer to the above method embodiments. The details will not be described again in this specification.

[0291] Optionally, memory 1102 may be read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, or random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or may be electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other compact disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disc storage media or other magnetic storage devices, or any other medium usable to carry or store expected program code in the form of instructions or data structures and accessible by a computer. However, this is not limited thereto. Memory 1102 may be integrated with processor 1101 or may exist separately and be coupled to processor 1101 through interface circuitry (not shown in FIG. 11 ) of communication device 1100. This is not particularly limited in this embodiment of this application.

[0292] The transceiver 1103 is configured to communicate with other communication devices. For example, the communication device 1100 is a terminal, and the transceiver 1103 may be configured to communicate with a network device or other terminal device. In another example, the communication device 1100 is a network device, and the transceiver 1103 may be configured to communicate with a terminal or other network device.

[0293] Optionally, the transceiver 1103 may include a receiver and a transmitter (not shown separately in FIG. 11), where the receiver is configured to implement a receiving function and the transmitter is configured to implement a transmitting function.

[0294] Optionally, the transceiver 1103 may be integrated with the processor 1101 or may exist independently and be coupled to the processor 1101 through an interface circuit (not shown in FIG. 11 ) of the communication device 1100, which is not particularly limited in this embodiment of the application.

[0295] It may be understood that the structure of the communications device 1100 shown in Figure 11 does not constitute a limitation on the communications device. An actual communications device may include more or fewer parts than those shown in the drawings, may combine some parts, or may have a different arrangement of parts.

[0296] Furthermore, the technical effects of the communication device 1100 refer to the technical effects of the method in the above method embodiments, and the details will not be described again herein.

[0297] It should be understood that the processor in the embodiments of this application may be a central processing unit (CPU), or the processor may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor may be any conventional processor, etc.

[0298] It may be further understood that the memory in the embodiments of this application may be volatile memory or nonvolatile memory, or may include volatile memory and nonvolatile memory. The nonvolatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory may be random access memory (RAM), and may be used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) may be used, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchlink dynamic random access memory (SLDRAM), and direct rambus random access memory (DR RAM).

[0299] All or part of the above embodiments may be realized by using software, hardware (e.g., circuits), firmware, or any combination thereof. When software is used to realize the embodiments, the above embodiments may be fully or partially realized in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded and executed on a computer, the procedures or functions according to the embodiments of this application are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer program or instructions may be stored in a computer-readable storage medium or transmitted from a computer-readable storage medium to another computer-readable storage medium. For example, the computer program or instructions may be transmitted from a website, computer, server, or data center to another website, computer, server, or data center via a wired (e.g., infrared, wireless, or microwave) method. The computer-readable storage medium may be any available medium accessible by a computer, or a data storage device such as a server or data center that integrates one or more available media. The media available may be magnetic media (e.g., floppy disks, hard disk drives, or magnetic tapes), optical media (e.g., DVDs), or semiconductor media, which may be solid-state drives.

[0300] It should be understood that the term "and / or" in this specification describes only the association relationship between related objects and represents that three relationships may exist. For example, A and / or B may represent the following three cases: only A exists, both A and B exist, and only B exists. A and B may be singular or plural. Furthermore, the character " / " in this specification usually represents an "or" relationship between related objects, but may also represent an "and / or" relationship. For more details, please refer to the context for understanding.

[0301] In this application, "at least one" means one or more, and "plurality" means two or more. "At least one of the following items" or similar phrases means any combination of those items, including any combination of singular or plural items. For example, at least one of a, b, or c may refer to a, b, c, a and b, a and c, b and c, or a, b, and c, where a, b, and c may be singular or plural.

[0302] It should be understood that the sequence numbers of the above processes do not mean the execution order in various embodiments of this application, and the execution order of the processes should be determined according to the functions and internal logic of the processes, and should not be construed as any limitation on the implementation process of the embodiments of this application.

[0303] Those skilled in the art may recognize that, in combination with the examples described in the embodiments disclosed in this specification, the units and algorithm steps may be realized by electronic hardware or a combination of computer software and electronic hardware. Whether a function is performed by hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to realize the described functions for each specific application, but the implementation method should not be considered to go beyond the scope of this application.

[0304] For the purpose of convenient and concise description, those skilled in the art can clearly understand that the detailed operation processes of the above systems, devices and units may refer to the corresponding processes in the above method embodiments, and the details will not be described again in this specification.

[0305] In some embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods may be implemented in other ways. For example, the described device embodiments are merely examples. For example, the division into units is merely a logical division of function, and other divisions may occur in actual implementations. For example, multiple units or components may be combined or integrated into other systems, or some features may be ignored or not implemented. Furthermore, the shown or discussed mutual couplings or direct couplings or communication connections may be realized through some interfaces. Indirect couplings or communication connections between devices or units may be realized in electronic, mechanical, or other forms.

[0306] The units described as separate parts may or may not be physically separate, and the parts shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of the units may be selected based on actual requirements to achieve the objectives of the solutions of the embodiments.

[0307] Furthermore, the functional units in the embodiments of this application may be integrated into one processing unit, each of the units may exist physically alone, or two or more units may be integrated into one unit.

[0308] When a function is realized in the form of a software functional unit and sold or used as an independent product, the function may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application may essentially be realized, or some of the parts contributing to the prior art or the technical solutions may be realized in the form of a software product. A computer software product is stored in a storage medium and includes some instructions for instructing a computer device (which may be a personal computer, a server, a network device, etc.) to execute all or some of the steps of the methods described in the embodiments of this application. The above storage medium includes any medium capable of storing program code, such as a USB flash drive, a removable hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0309] The above description is merely a specific implementation of this application and is not intended to limit the scope of protection of this application. Any variations or replacements that can be easily conceived by those skilled in the art within the technical scope disclosed in this application shall fall within the scope of protection of this application. Therefore, the scope of protection of this application shall be subject to the scope of protection of the claims.

Claims

1. 1. A communication method comprising: Measuring a reference signal (RS) from a network device by using a first RS resource set to obtain a measurement result of the RS; inputting the measurement results into a neural network model to determine a target RS resource set, where the target RS resource set is determined from candidate RS resource sets, and the candidate RS resource set is determined by the neural network model based on the measurement results, where the candidate RS resource set belongs to a second RS resource set, and an RS transmitted by using the first RS resource set is the same as or different from an RS transmitted by using the second RS resource set; A method comprising:

2. 2. The method of claim 1, wherein the target RS resource set is determined from the candidate RS resource sets based on a first threshold and information about the candidate RS resource sets, the first threshold being related to the information about the candidate RS resource sets.

3. The method of claim 2, wherein the candidate RS resource set is determined by the neural network model based on the measurement results, specifically, the information about the candidate RS resource set is determined by the neural network model based on the measurement results.

4. 4. The method of claim 3, wherein the information about the candidate RS resource sets includes at least one of the following: a probability that each RS resource in the candidate RS resource set is an optimal RS resource; a signal quality of each RS resource in the candidate RS resource set; or an angle of each RS resource in the candidate RS resource set, wherein the angle of each RS resource in the candidate RS resource set is an angle difference between a transmit beam and a receive beam corresponding to the RS resource.

5. 5. The method of claim 4, wherein the information about the target RS resource set and the first threshold satisfy at least one of the following relationships: the probability that an RS resource in the target RS resource set is an optimal RS resource is greater than the probability represented by the first threshold; the signal quality of the RS resource in the target RS resource set is greater than the signal quality represented by the first threshold; or the angle of each RS resource in the RS resource set is less than the angle represented by the first threshold.

6. The method of claim 2 , further comprising the step of receiving indication information from the network device, the indication information indicating the first threshold value.

7. The method of claim 2 , wherein the first threshold is pre-configured in the neural network model.

8. The method of claim 2 , further comprising the step of: transmitting the information about the target RS resource set to the network device.

9. 9. The method of claim 8, wherein the information about the target RS resource set includes at least one of the following: an identifier of an RS resource in the target RS resource set; a signal quality of an RS resource in the target RS resource set; or an angle of each RS resource in the RS resource set, wherein the angle of each RS resource in the target RS resource set is an angle difference between a transmit beam and a receive beam corresponding to the RS resource.

10. The method of any one of claims 1 to 9, wherein there are multiple target RS resource sets, the multiple target RS resource sets being configured for use in different time units.

11. 1. A communication method comprising: transmitting a reference signal (RS) to a terminal by using a first RS resource set; receiving measurement results of the RS fed back by the terminal; inputting the measurement results into a neural network model to determine a target RS resource set, where the target RS resource set is determined from candidate RS resource sets, and the candidate RS resource set is determined by the neural network model based on the measurement results, where the candidate RS resource set belongs to a second RS resource set, and an RS transmitted by using the first RS resource set is the same as or different from an RS transmitted by using the second RS resource set; A method comprising:

12. 12. The method of claim 11, wherein the target RS resource set is determined from the candidate RS resource sets based on a first threshold and information about the candidate RS resource sets, the first threshold being related to the information about the candidate RS resource sets.

13. The method of claim 12, wherein the candidate RS resource set is determined by the neural network model based on the measurement results, specifically, the information about the candidate RS resource set is determined by the neural network model based on the measurement results.

14. 14. The method of claim 13, wherein the information about the candidate RS resource sets includes at least one of the following: a probability that each RS resource in the candidate RS resource set is an optimal RS resource; a signal quality of each RS resource in the candidate RS resource set; or an angle of each RS resource in the candidate RS resource set, wherein the angle of each RS resource in the candidate RS resource set is an angle difference between a transmit beam and a receive beam corresponding to the RS resource.

15. 15. The method of claim 14, wherein the information about the target RS resource set and the first threshold satisfy at least one of the following relationships: the probability that an RS resource in the target RS resource set is an optimal RS resource is greater than the probability represented by the first threshold; the signal quality of the RS resource in the target RS resource set is greater than the signal quality represented by the first threshold; or the angle of each RS resource in the RS resource set is less than the angle represented by the first threshold.

16. 16. The method of claim 12, wherein the first threshold is pre-configured in the neural network model.

17. The method of claim 11 , further comprising transmitting the target RS resource set to the terminal.

18. 18. The method of claim 11, wherein there are multiple target RS resource sets, the multiple target RS resource sets being configured for use in different time units.

19. 19. The method according to any one of claims 11 to 18, wherein the measurements of the RS are all measurements or a subset of measurements.

20. A communication device, Apparatus comprising modules adapted to carry out the method of any one of claims 1 to 10.

21. A communication device, 20. Apparatus comprising modules configured to carry out the method of any one of claims 11 to 19.

22. A communication device, 20. A communications device comprising at least one processor and a memory, the memory configured to store computer instructions that, when executed by the at least one processor, enable the communications device to perform a method according to any one of claims 11 to 19.

23. 1. A communication system comprising:

20. A communication system comprising a terminal and a network device, the terminal being configured to perform the method of any one of claims 1 to 10 and the network device being configured to perform the method of any one of claims 11 to 19.

24. 1. A computer-readable storage medium, comprising:

20. The computer readable storage medium contains a computer program or instructions which, when executed on a computer, enable the computer to carry out the method of any one of claims 1 to 19.

Images