Verification method and apparatus, and device and storage medium

Abstract

A verification method and apparatus, and a device and a storage medium, which relate to the technical field of communications. The method comprises: a first device obtaining first channel information on the basis of first reference signals of M1 ports, wherein the first channel information is channel information corresponding to N ports, N being an integer greater than 1, and M1 being a positive integer less than N (410); the first device obtaining second channel information on the basis of second reference signals of M2 ports, wherein the second channel information is channel information corresponding to the M2 ports, M2 being a positive integer less than or equal to N (420); and the first device verifying the first channel information on the basis of the second channel information (430). The method implements the verification of the accuracy / validity of predicted channel information.

Description

Verification methods, apparatus, equipment and storage media Technical Field

[0001] This application relates to the field of communication technology, and in particular to a verification method, apparatus, device, and storage medium. Background Technology

[0002] In traditional CSI (Channel State Information) feedback schemes, to obtain complete uplink or downlink CSI, the transmitter needs to send a complete antenna port (usually corresponding to the number of antennas at the transmitter) and a reference signal corresponding to the complete bandwidth. The receiver measures the above reference signal to obtain complete channel information, thereby obtaining the uplink or downlink CSI and indicating it to the transmitter.

[0003] When there are many antennas and a large number of antenna ports at the transmitting end (such as 128 / 256 downlink ports and 16 uplink ports), and the CSI measurement bandwidth is large, frequently sending the reference signal corresponding to the complete antenna port and the complete bandwidth requires a large amount of reference signal resources, which will affect the uplink and downlink data transmission rates.

[0004] By leveraging AI (Artificial Intelligence) / ML (Machine Learning) technologies, the Channel Indicator (CSI) over the entire antenna port or bandwidth can be predicted from channel information obtained from a portion of the antenna port or bandwidth, thereby reducing the overhead of the reference signal. However, how to verify the accuracy or effectiveness of the predicted channel information using this approach requires further research.

[0005] Summary of the Invention

[0006] This application provides a verification method, apparatus, device, and storage medium. The technical solutions provided by this application are as follows.

[0007] According to one aspect of the embodiments of this application, a verification method is provided, the method being executed by a first device, the method comprising:

[0008] Based on the first reference signals of M1 ports, the first channel information is obtained. The first channel information is the channel information corresponding to N ports, where N is an integer greater than 1 and M1 is a positive integer less than N.

[0009] Based on the second reference signals of the M2 ports, the second channel information is obtained. The second channel information is the channel information corresponding to the M2 ports, where M2 is a positive integer less than or equal to N.

[0010] The first channel information is verified based on the second channel information.

[0011] According to one aspect of the embodiments of this application, a verification method is provided, the method being performed by a second device, the method comprising:

[0012] Send first reference signals for M1 ports, wherein the first reference signals for M1 ports are used to determine first channel information, wherein the first channel information is channel information corresponding to N ports, where N is an integer greater than 1 and M1 is a positive integer less than N;

[0013] Send second reference signals for M2 ports, the second reference signals for M2 ports are used to determine second channel information, the second channel information is the channel information corresponding to the M2 ports, and M2 is a positive integer less than or equal to N;

[0014] The second channel information is used to verify the first channel information.

[0015] According to one aspect of the embodiments of this application, a verification apparatus is provided, the apparatus comprising:

[0016] The processing module is used to obtain first channel information based on the first reference signals of M1 ports. The first channel information is the channel information corresponding to N ports, where N is an integer greater than 1 and M1 is a positive integer less than N.

[0017] The processing module is further configured to obtain second channel information based on the second reference signals of the M2 ports, wherein the second channel information is the channel information corresponding to the M2 ports, and M2 is a positive integer less than or equal to N;

[0018] The processing module is further configured to verify the first channel information based on the second channel information.

[0019] According to one aspect of the embodiments of this application, a verification apparatus is provided, the apparatus comprising:

[0020] The transmitting module is used to transmit first reference signals for M1 ports. The first reference signals for M1 ports are used to determine first channel information. The first channel information is the channel information corresponding to N ports, where N is an integer greater than 1 and M1 is a positive integer less than N.

[0021] The transmitting module is further configured to transmit second reference signals for M2 ports, the second reference signals for M2 ports being used to determine second channel information, the second channel information being the channel information corresponding to the M2 ports, where M2 is a positive integer less than or equal to N;

[0022] The second channel information is used to verify the first channel information.

[0023] According to one aspect of the embodiments of this application, a communication device is provided, the communication device including a processor and a memory, the memory storing a computer program, the processor executing the computer program to implement the verification method executed by the first device or the verification method executed by the second device.

[0024] According to one aspect of the embodiments of this application, a computer-readable storage medium is provided, the storage medium storing a computer program, the computer program being executed by a processor to implement the verification method executed by the first device or the verification method executed by the second device.

[0025] According to one aspect of the embodiments of this application, a chip is provided, the chip including programmable logic circuits and / or program instructions, which, when the chip is running, are used to implement the verification method executed by the first device or the verification method executed by the second device.

[0026] According to one aspect of the embodiments of this application, a computer program product is provided, the computer program product including computer instructions stored in a computer-readable storage medium, wherein a processor reads from the computer-readable storage medium and executes the computer instructions to implement the verification method executed by the first device or the verification method executed by the second device.

[0027] The technical solutions provided in this application embodiment may have the following beneficial effects:

[0028] For scenarios where full-port channel information is predicted based on reference signal measurements from a subset of ports, the accuracy and validity of the predicted channel information can be verified by comparing the actual measured channel information with the predicted channel information. In scenarios where full-port channel information is predicted using an AI / ML model, the above approach enables the monitoring and evaluation of the AI / ML model's performance. Attached Figure Description

[0029] Figure 1 is a schematic diagram of the implementation environment of a solution provided in an embodiment of this application;

[0030] Figure 2 is a schematic diagram of a neural network structure provided in an embodiment of this application;

[0031] Figure 3 is a schematic diagram of different periodic CSI reporting methods provided in one embodiment of this application;

[0032] Figure 4 is a flowchart of a verification method provided in an embodiment of this application;

[0033] Figure 5 is a flowchart of a method for performance monitoring of an AI / ML model provided in an embodiment of this application;

[0034] Figure 6 is a schematic diagram of M1 ports and M2 ports provided in one embodiment of this application;

[0035] Figure 7 is a flowchart of a method for performance monitoring of AI / ML models provided in another embodiment of this application;

[0036] Figure 8 is a flowchart of a method for performance monitoring of AI / ML models provided in another embodiment of this application;

[0037] Figure 9 is a flowchart of a verification method provided in another embodiment of this application;

[0038] Figure 10 is a schematic diagram of the AI / ML model input under the method shown in Figure 9;

[0039] Figure 11 is a flowchart of a method for performance monitoring of AI / ML models provided in another embodiment of this application;

[0040] Figure 12 is a flowchart of a method for performance monitoring of AI / ML models provided in another embodiment of this application;

[0041] Figure 13 is a flowchart of a method for performance monitoring of AI / ML models provided in another embodiment of this application;

[0042] Figure 14 is a block diagram of a verification device provided in an embodiment of this application;

[0043] Figure 15 is a block diagram of a verification device provided in another embodiment of this application;

[0044] Figure 16 is a block diagram of a communication device provided in one embodiment of this application. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0046] The network architecture and business scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0047] The technical solutions of this application embodiment can be applied to various communication systems, such as: Global System for Mobile communication (GSM) system, Code Division Multiple Access (CDMA) system, Wideband Code Division Multiple Access (WCDMA) system, General Packet Radio Service (GPRS), Long Term Evolution (LTE) system, Advanced Long Term Evolution (LTE-A) system, New Radio (NR) system, evolution system of NR system, LTE-based access to unlicensed spectrum (LTE-U) system, NR-based access to unlicensed spectrum (NR-U) system, Non-Terrestrial Networks (NTN) system, Universal Mobile Telecommunication System (UMTS), Wireless Local Area Networks (WLAN), Wireless Fidelity (WiFi), and 5G (5G) communication. th -Generation, 5G) system, B5G (Beyond 5G) system, sixth-generation communication (6 th -Generation, 6G) systems or other communication systems, etc.

[0048] Traditional communication systems typically support a limited number of connections and are easy to implement. However, with the development of communication technology, mobile communication systems will not only support traditional communication but also, for example, device-to-device (D2D) communication, machine-to-machine (M2M) communication, machine-type communication (MTC), vehicle-to-vehicle (V2V) communication, or vehicle-to-everything (V2X) communication. The embodiments of this application can also be applied to these communication systems.

[0049] The communication system in this application embodiment can be applied to carrier aggregation (CA) scenarios, dual connectivity (DC) scenarios, and standalone (SA) network deployment scenarios.

[0050] The communication system in this application embodiment can be applied to unlicensed spectrum, wherein unlicensed spectrum can also be considered as shared spectrum; or, the communication system in this application embodiment can also be applied to licensed spectrum, wherein licensed spectrum can also be considered as non-shared spectrum.

[0051] The embodiments of this application can be applied to both non-terrestrial networks (NTN) and terrestrial networks (TN). NTN typically uses satellite communication to provide communication services to terrestrial users. Currently, NTN systems include NR-NTN and IoT-NTN systems, and other NTN systems may be included in the future.

[0052] Please refer to Figure 1, which shows a schematic diagram of a network architecture 100 provided in one embodiment of this application. The network architecture 100 may include: a terminal device 10, an access network device 20, and a core network element 30.

[0053] Terminal device 10 may refer to UE (User Equipment), access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, wireless communication device, user agent, or user equipment. In some embodiments, terminal device 10 may also be a cellular phone, cordless phone, SIP (Session Initiation Protocol) phone, WLL (Wireless Local Loop) station, PDA (Personal Digital Assistant), handheld device with wireless communication capabilities, computing device or other processing device connected to a wireless modem, vehicle-mounted device, wearable device, 5GS (5 thTerminal devices in a Generation System (5G mobile communication system) or in a future evolved PLMN (Public Land Mobile Network), etc., are not limited to this embodiment. For ease of description, the devices mentioned above are collectively referred to as terminal devices. The number of terminal devices 10 is usually multiple, and one or more terminal devices 10 can be distributed within the cell managed by each access network device 20. Terminal devices can also be simply referred to as terminals or UEs, the meaning of which will be understood by those skilled in the art.

[0054] Access network device 20 is a device deployed in an access network to provide wireless communication functionality to terminal device 10. Access network device 20 may include various forms of macro base stations, micro base stations, relay stations, access points, etc. In systems employing different wireless access technologies, the name of the device with access network device functionality may differ; for example, in a 5G NR system, it is called gNodeB or gNB. As communication technologies evolve, the name "access network device" may change. For ease of description, in this embodiment, the aforementioned devices providing wireless communication functionality to terminal device 10 are collectively referred to as access network devices. In some embodiments, a communication relationship can be established between terminal device 10 and core network element 30 through access network device 20. For example, in an LTE (Long Term Evolution) system, access network device 20 may be one or more eNodeBs in an EUTRAN (Evolved Universal Terrestrial Radio Access Network) or EUTRAN; in a 5G NR system, access network device 20 may be one or more gNBs in a RAN (Radio Access Network). In the embodiments of this application, unless otherwise specified, the term "network device" refers to access network device 20, such as a base station.

[0055] Core network element 30 is a network element deployed in the core network. Its main functions are to provide user connectivity, manage users, and bear services, serving as an interface to external networks. For example, core network elements in a 5G NR system may include AMF (Access and Mobility Management Function) entities, UPF (User Plane Function) entities, and SMF (Session Management Function) entities.

[0056] In some embodiments, the access network device 20 and the core network element 30 communicate with each other via some air interface technology, such as the NG interface in a 5G NR system. The access network device 20 and the terminal device 10 communicate with each other via some air interface technology, such as the Uu interface.

[0057] The "5G NR system" in this application embodiment can also be referred to as a 5G system or an NR system, but those skilled in the art will understand its meaning. The technical solutions described in this application embodiment can be applied to LTE systems, 5G NR systems, and subsequent evolution systems of 5G NR systems (such as B5G (Beyond 5G, a fifth-generation mobile communication technology) systems, 6G systems (6G... th The sixth-generation mobile communication system can also be applied to other communication systems such as NB-IoT (Narrow Band Internet of Things) systems, but this application does not limit it.

[0058] In this embodiment, the network device can provide services to a cell. The terminal device communicates with the network device through the transmission resources (e.g., frequency domain resources, or spectrum resources) on the carrier used by the cell. The cell can be the cell corresponding to the network device (e.g., a base station). The cell can belong to a macro base station or to a base station corresponding to a small cell. The small cell can include: metro cell, micro cell, pico cell, femto cell, etc. These small cells have the characteristics of small coverage area and low transmission power, and are suitable for providing high-speed data transmission services.

[0059] Before introducing the technical solution of this application, some related technical knowledge involved in this application will be introduced and explained. The following related technologies are optional solutions and can be arbitrarily combined with the technical solutions of the embodiments of this application, all of which fall within the protection scope of the embodiments of this application. The embodiments of this application include at least some of the following contents.

[0060] 1. Neural Networks (NNs) and Machine Learning

[0061] A neural network is a computational model consisting of multiple interconnected neurons. The neuron structure is shown in sub-Figure 1 of Figure 2, where a1, ..., an, 1 are input signals, and w1, ..., wn, b are weights. The connections between nodes represent the weighted sums from the input signals to the output signals, and are called weights. Each node performs a weighted summation of different input signals and obtains the output t through a specific activation function f.

[0062] A simple neural network, as shown in sub-Figure 2 of Figure 2, includes an input layer, hidden layers, and an output layer. Through different connections, weights, and activation functions of multiple neurons, different outputs can be generated, thus fitting a mapping relationship from input to output. Each higher-level node is connected to all its lower-level nodes. This fully connected model can also be called a DNN (Deep Neural Network) in this application and can be used as the NN model in this application.

[0063] The basic structure of a CNN (Convolutional Neural Network) is shown in sub-Figure 3 of Figure 2. It can include an input layer, multiple convolutional layers, multiple pooling layers, fully connected layers, and an output layer. Each neuron in the convolutional kernel of the convolutional layer is locally connected to its input. By introducing pooling layers, the maximum or average value features of a certain local area are extracted, effectively reducing the network's parameters and mining local features. This allows the convolutional neural network to converge quickly and achieve excellent performance.

[0064] Recurrent Neural Networks (RNNs) are neural networks that model sequential data and have achieved remarkable success in natural language processing applications such as machine translation and speech recognition. Specifically, the network memorizes information from past time steps and uses it in the calculation of the current output. That is, nodes in the hidden layers are no longer disconnected but connected, and the input to the hidden layers includes not only the input layer but also the output of the hidden layer from the previous time step. Commonly used RNN structures include Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU).

[0065] 2. Downlink CSI Feedback

[0066] To enable network devices to perform reasonable scheduling, terminal devices need to report downlink CSI (Channel State Information) so that the base station can determine scheduling information for the terminal devices, such as the transmission layer number, precoding matrix, transmission beam, and modulation / coding scheme. Specifically, the terminal device's CSI reporting is based on the CSI reporting configuration indicated by the network device and the CSI-RS (Channel State Information-Reference Signal) sent by the network device. The uplink resources used by the terminal device for CSI reporting and the CSI-RS used for CSI measurement are both indicated by the CSI reporting configuration. Each CSI reporting configuration corresponds to one CSI report, and each CSI report can include different information such as CRI (CSI-RS Resource Indicator), RI (Rank Indicator), PMI (Precoding Matrix Indicator), and CQI (Channel Quality Indicator). This information is obtained based on the CSI-RS configured and sent by the network device. Specifically, the content / information included in the CSI is determined by the report quantity information in the CSI reporting configuration. This reporting volume information can indicate at least one of the following reporting volumes.

[0067] CRI is used to determine the CSI-RS resource currently used for channel measurement and the Interference Measurement Resource (IMR) currently used for interference measurement from multiple CSI-RS resources.

[0068] RI is used to report the recommended number of transport layers.

[0069] PMI is used to determine the recommended precoding matrix from a predefined codebook.

[0070] CQI is used to report the current channel quality.

[0071] RSRP (Reference Signal Receiving Power) is used to report the RSRP of the SSB (Synchronous Signal Block) or CSI-RS corresponding to the fed-back index, so that the network side can determine the beam used for downlink transmission.

[0072] LI is used to report the index of the transport layer associated with the phase tracking reference signal.

[0073] RI / PMI / CQI can be determined based on the SINR (Signal to Interference plus Noise Ratio) estimated by the terminal equipment. The channel component of SINR is determined based on a non-zero power CSI-RS configured in the network for channel measurement, while the interference component is determined based on a CSI-IM or a non-zero power CSI-RS configured in the network for interference measurement. The CSI-RS resources used for channel measurement can include multiple antenna ports to measure the complete downlink channel and thus calculate the CSI.

[0074] Terminal devices can report CSI in three periodic ways: periodic CSI, quasi-persistent CSI, and aperiodic CSI, as shown in Figure 3. Periodic CSI is transmitted on the PUCCH (Physical Uplink Control Channel), and its CSI reporting configuration is configured by the RRC (Radio Resource Control). After receiving the corresponding RRC configuration, the terminal device periodically reports the CSI. Quasi-persistent CSI can be transmitted on either the PUCCH or PUSCH (Physical Uplink Shared Channel). The CSI reporting configuration corresponding to CSI transmitted on the PUCCH is pre-configured by RRC signaling and activated or deactivated by MAC (Medium Access Control) layer signaling. The CSI reporting configuration corresponding to CSI transmitted on the PUSCH is dynamically indicated (activated or deactivated) by DCI (Downlink Control Information) signaling. After receiving activation or indication signaling from the network configuration, the terminal device periodically transmits CSI on the PUCCH or PUSCH until it receives deactivation signaling and stops reporting. The CSI reporting configuration for non-periodic CSI reporting is also pre-configured via RRC signaling. Part of the configuration can be activated via MAC layer signaling, and the CSI reporting configuration used for CSI reporting can be indicated via CSI trigger signaling in the DCI. Upon receiving the CSI trigger signaling, the terminal device reports the corresponding CSI on the scheduled PUSCH in one go according to the indicated CSI reporting configuration.

[0075] Please refer to Figure 4, which shows a flowchart of a verification method provided in one embodiment of this application. This method can be applied to the network architecture shown in Figure 1. As shown in Figure 4, the method may include at least one of the following steps 410 to 430.

[0076] Step 410: The first device obtains the first channel information based on the first reference signals of M1 ports. The first channel information is the channel information corresponding to N ports, where N is an integer greater than 1 and M1 is a positive integer less than N.

[0077] In some embodiments, the first reference signal is transmitted by the second device, which transmits the first reference signal across M1 ports; or, in other words, the second device transmits the first reference signal across M1 ports. In this embodiment, a port can also be referred to as an antenna port, and each antenna port corresponds to an antenna radio frequency unit.

[0078] In some embodiments, the first device and the second device are two ends of communication. Exemplarily, the first device is a terminal device, and the second device is a network device (such as a base station). Exemplarily, the first device is a network device (such as a base station), and the second device is a terminal device. Exemplarily, the first device is a first terminal device, and the second device is a second terminal device; the first terminal device and the second terminal device are two different terminal devices.

[0079] In some embodiments, the maximum configurable number of ports on the second device side is N, meaning the second device can transmit reference signals from a maximum of N ports. In step 410, the second device transmits the first reference signal on a subset of the N ports (i.e., M1 ports, M1 < N). Compared to the second device transmitting the first reference signal to the first device on N ports, this embodiment reduces the number of ports used to transmit the first reference signal, thereby reducing the overhead of transmitting the first reference signal between the first and second devices.

[0080] In some embodiments, when the first reference signal is a downlink reference signal, the second device is a network device and the first device is a terminal device. When the first reference signal is a downlink reference signal, the first reference signal can be CSI-RS, and the first device can feed back one or more of CRI, RI, PMI, CQI, RSRP, and LI information in the CSI. Of course, the first reference signal can also be other downlink reference signals, such as DMRS (Demodulation Reference Signal), etc., and this application does not limit it to these.

[0081] Furthermore, when the first reference signal is an uplink reference signal, the second device is a terminal device and the first device is a network device. When the first reference signal is an uplink reference signal, the first reference signal can be DMRS, SRS (Sounding Reference Signal), etc., and this application does not limit this. When the first reference signal is a sidelink reference signal, the second device is a second terminal device and the first device is a first terminal device. When the first reference signal is a sidelink reference signal, the first reference signal can be SL-PRS (Sidelink Positioning Reference Signal), SL-DMRS (Sidelink Demodulation Reference Signal), etc., and this application does not limit this.

[0082] In some embodiments, the first device measures the first reference signals of M1 ports to obtain third channel information. This third channel information is the channel information corresponding to the M1 ports obtained from the measurement of the first reference signals of the M1 ports. The first device then obtains first channel information based on the third channel information. The first channel information is the channel information corresponding to N ports; it can also be understood as the channel information corresponding to all ports of the second device. Through this method, the channel information corresponding to all ports is obtained based on the channel information corresponding to a subset of ports.

[0083] In some embodiments, the first channel information is output by an AI / ML model. The input to the AI / ML model includes third channel information, which is the channel information corresponding to the M1 ports obtained by measuring the first reference signal of the M1 ports. The AI / ML model can be a model built on a neural network, which has the function of predicting the channel information corresponding to all ports based on the channel information corresponding to some ports. For example, the third channel information is input into the AI / ML model, and the AI / ML model outputs the first channel information based on the third channel information.

[0084] Step 420: The first device obtains the second channel information based on the second reference signals of the M2 ports. The second channel information is the channel information corresponding to the M2 ports, where M2 is a positive integer less than or equal to N.

[0085] In some embodiments, the second reference signal is transmitted by a second device, which transmits the second reference signal across M2 ports. The second reference signal and the first reference signal can be the same type of reference signal or two different reference signals; this application does not limit this. For example, the second reference signal and the first reference signal are the same type of reference signal, both being CSI-RS.

[0086] In step 420, one possible implementation is that the second device sends the first reference signal on a portion of the N ports (i.e., M2 ports, M2 < N).

[0087] In step 420, another possible implementation is that the second device sends the first reference signal on all of the N ports (i.e., M2 ports, M2=N).

[0088] In some embodiments, the first device measures the second reference signal of M2 ports to obtain second channel information, which is the channel information corresponding to the M2 ports obtained by measuring the second reference signal of the M2 ports.

[0089] In some embodiments, the value of N can be 32, 64, 128, 256, 512, 1024, 2048, 4096, etc., and this application does not limit this. M1 is less than N, M2 can be less than N, or M2 can be equal to N. For example, it can be calculated according to 1 / (2) of the value of N. m Sampling is performed to obtain M1 or M2 ports, where m is a positive integer.

[0090] Step 430: The first device verifies the first channel information based on the second channel information.

[0091] Since the second channel information is the channel information actually measured by the first device, while the first channel information is the channel information predicted by the first device, such as the channel information predicted by an AI / ML model, the accuracy / validity of the first channel information can be verified based on the second channel information.

[0092] In some embodiments, the channel information described above may be a channel measured based on a reference signal, or it may be a feature vector. Alternatively, if the channel information is CSI, the CSI may at least include PMI information. The PMI information may be used by the first device to obtain the precoding matrix used for transmission. Optionally, the CSI may also include other information such as RI and CQI.

[0093] In some embodiments, the first device compares the second channel information with the channel information of the corresponding M2 ports in the first channel information to obtain a comparison result, and verifies the first channel information based on the comparison result. When M2 is less than N, the channel information corresponding to the M2 ports is extracted from the channel information corresponding to the N ports included in the first channel information, and then the second channel information is compared with the channel information corresponding to the M2 ports extracted from the first channel information to obtain a comparison result. For example, if N = 32, M2 = 4, the indices of the N ports are 0 to 31, and the indices of the M2 ports are 0, 8, 16, and 24, then the channel information corresponding to the four ports with indices 0, 8, 16, and 24 are extracted from the first channel information and compared with the second channel information. When M2 equals N, the second channel information is directly compared with the first channel information to obtain a comparison result.

[0094] In some embodiments, comparing the second channel information with the channel information corresponding to the M2 ports extracted from the first channel information is to compare the similarity / closeness between the two information.

[0095] In some embodiments, the comparison result is used to characterize the SGCS (Squared Generalized Cosine Similarity) between the second channel information and the channel information of the corresponding M2 ports in the first channel information; or, the comparison result is used to characterize the difference between the second channel information and the channel information of the corresponding M2 ports in the first channel information; or, the comparison result is used to characterize the correlation between the second channel information and the channel information of the corresponding M2 ports in the first channel information. That is, the two pieces of information can be compared based on indicators such as SGCS, difference, and correlation to obtain the degree of similarity / closeness between the two pieces of information. Of course, this application does not limit the use of other indicators to measure the degree of similarity / closeness between the two pieces of information, such as GCS (Generalized Cosine Similarity), BLER (Block Error Rate), Spectral Efficiency, etc.

[0096] In some embodiments, if the comparison result meets a first condition, the first channel information is determined to be accurate / valid, and / or the AI / ML model is determined to be accurate / valid. In some embodiments, if the comparison result does not meet the first condition, the first channel information is determined to be inaccurate / invalid, and / or the AI / ML model is determined to be inaccurate / invalid. The first condition relates to the metrics used in the comparison result.

[0097] For example, the metric used for comparison is SGCS. If the SGCS between the channel information of the M2 ports corresponding to the second channel information and the first channel information is greater than or not less than a first threshold, the first channel information is determined to be accurate / valid, and / or the AI / ML model is determined to be accurate / valid. If the SGCS between the channel information of the M2 ports corresponding to the second channel information and the first channel information is not greater than or less than the first threshold, the first channel information is determined to be inaccurate / invalid, and / or the AI / ML model is determined to be inaccurate / invalid. The value of SGCS can be in the range [0,1]. The closer the SGCS is to 1, the more accurate the first channel information and the better the performance of the AI / ML model.

[0098] For example, the comparison result uses the difference as the metric. If the difference between the channel information of the M2 corresponding ports in the second channel information and the first channel information is less than or not greater than a second threshold, the first channel information is determined to be accurate / valid, and / or the AI / ML model is determined to be accurate / valid. If the difference between the channel information of the M2 corresponding ports in the second channel information and the first channel information is not less than or greater than the second threshold, the first channel information is determined to be inaccurate / invalid, and / or the AI / ML model is determined to be inaccurate / invalid. That is, the smaller the difference, the more accurate the first channel information, and the better the performance of the AI / ML model.

[0099] For example, the metric used for comparison is correlation. If the correlation between the second channel information and the channel information of the corresponding M² ports in the first channel information is greater than or not less than a third threshold, the first channel information is determined to be accurate / valid, and / or the AI / ML model is determined to be accurate / valid. If the correlation between the second channel information and the channel information of the corresponding M² ports in the first channel information is not greater than or not less than the third threshold, the first channel information is determined to be inaccurate / invalid, and / or the AI / ML model is determined to be inaccurate / invalid. That is, the greater the correlation, the more accurate the first channel information, and the better the performance of the AI / ML model.

[0100] In addition, the aforementioned first threshold, second threshold, and third threshold may be agreed upon by the protocol, pre-configured, or configured by the network device; this application does not limit this.

[0101] It should be understood that, in this application, verifying the first channel information can be understood as verifying the accuracy of the first channel information, or as verifying the effectiveness of the AI / ML model. Verification can also be replaced by monitoring, performance monitoring, effectiveness monitoring, etc.

[0102] The technical solution provided in this application, for scenarios where channel information for the entire port is predicted based on reference signal measurement results from a portion of the port, allows for verification of the accuracy / validity of the predicted channel information by comparing the actually measured channel information with the predicted channel information. In scenarios where AI / ML models are used to predict channel information for the entire port, the above solution enables the monitoring and evaluation of the AI / ML model's performance.

[0103] The following section will introduce and explain several possible scenarios for transmitting the first reference signal and the second reference signal.

[0104] Case 1

[0105] In some embodiments, the first reference signal and the second reference signal are transmitted at different times.

[0106] Assume the moment when the second device sends the first reference signal is time 1, and the moment when it sends the second reference signal is time 2. Let time 1 be denoted as t1 and time 2 as t2. In case 1, t1 and t2 are different. For example, t2 is after t1.

[0107] In some embodiments, as shown in Figure 5, the method for performance monitoring of AI / ML models includes the following steps:

[0108] 1. The second device sends the first reference signal for M1 ports at time t1;

[0109] 2. The first device obtains the first channel information through an AI / ML model based on the measurement results of the first reference signals at M1 ports;

[0110] 3. After the performance monitoring of the AI / ML model is triggered, the second device sends the second reference signal of M2 ports at time t2;

[0111] 4. The first device measures the second reference signal of M2 ports to obtain the second channel information;

[0112] 5. The first device extracts the channel information corresponding to M2 ports from the first channel information, compares it with the second channel information, and determines whether the AI / ML model is accurate / effective based on the comparison result.

[0113] In case 1, the M1 ports and M2 ports can be obtained in any of the following ways.

[0114] In some embodiments, M1 ports and / or M2 ports are obtained from N ports through beamforming. Beamforming, also known as beamforming or spatial filtering, is a signal processing technique that uses a sensor array to transmit and receive signals in a directional manner. By adjusting the parameters of the basic units of the phase array, signals at certain angles achieve constructive interference, while signals at other angles achieve destructive interference, thereby achieving directional transmission and reception of signals. Through beamforming, M1 ports and / or M2 ports can be determined from N ports.

[0115] In some embodiments, M1 ports and / or M2 ports are obtained by sampling N ports in the horizontal and / or vertical dimensions. The N ports can be arranged in an array, so M1 ports can be obtained by sampling N ports in the horizontal and / or vertical dimensions, and similarly, M2 ports can be obtained by sampling N ports in the horizontal and / or vertical dimensions.

[0116] In some embodiments, M1 ports are the first M1 ports out of N ports, and / or M2 ports are the first M2 ports out of N ports. In this case, some of the M1 ports and M2 ports may be the same, or even completely identical.

[0117] In some embodiments, M1 ports are the last M1 ports out of N ports, and / or M2 ports are the last M2 ports out of N ports. In this case, some of the M1 ports and M2 ports may be the same, or even completely identical.

[0118] In some embodiments, the M1 ports and / or M2 ports are obtained by uniformly sampling from N ports. For example, N = 64, M1 = 4, M2 = 8, the indices of the N ports are 0 to 63, the indices of the M1 ports are 0, 16, 32, and 48, and the indices of the M2 ports are 1, 9, 17, 25, 33, 41, 49, and 57.

[0119] In some embodiments, M1 ports and / or M2 ports are sampled from n port groups obtained by dividing N ports, where n is a positive integer. For example, N = 64, the indices of the N ports are 0 to 63, and these N ports are divided into 4 port groups. Port group 0 contains 16 ports with indices 0 to 15, port group 1 contains 16 ports with indices 16 to 31, port group 2 contains 16 ports with indices 32 to 47, and port group 3 contains 16 ports with indices 48 to 63. Assuming M1 = 4, one port is sampled from each of the above 4 port groups to obtain M1 ports, for example, the indices of these M1 ports are 1, 20, 42, and 55. Assuming M2 = 8, two ports are sampled from each of the above 4 port groups to obtain M2 ports, for example, the indices of these M2 ports are 1, 7, 17, 23, 33, 39, 49, and 55.

[0120] In case 1, the M1 ports and the M2 ports are identical; or, the M1 ports and the M2 ports are completely different; or, the M1 ports and the M2 ports are partially identical. For example, as shown in Figure 6, the second device can be configured with a maximum of N ports, which are distributed in a 4×4 array. The M1 ports are the four ports in the first row, first column, first row, third column, third row, first column, and third row, third column, and the M2 ports are the four ports in the second row, second column, second row, fourth column, fourth row, second column, and fourth row, fourth column.

[0121] In scenario 1, model performance monitoring uses reference signals from a portion of the ports, which reduces the reference signal overhead during model performance monitoring. However, since the second time step occurs after the first time step, there is also the issue that the first channel information output by the AI / ML model may be outdated and cannot be used as a performance monitoring reference.

[0122] Case 2

[0123] In some embodiments, the first reference signal and the second reference signal are transmitted at the same time; or, the first reference signal and the second reference signal are transmitted in the same time unit.

[0124] Assume the time when the second device sends the first reference signal is time 1, and the time when it sends the second reference signal is time 2. Let time 1 be denoted as t1 and time 2 as t2. In case 2, t1 and t2 are the same, or t1 and t2 belong to the same time unit. Optionally, the time unit includes, but is not limited to, any of the following: time slot, subframe, mini-time slot, sub-time slot, or one or more symbols.

[0125] Optionally, the first reference signal and the second reference signal have the same timestamp. The timestamp can be represented by a value calculated in time units (e.g., milliseconds, microseconds, etc.); or, the timestamp can be represented by an SFN (System Frame Number) value, a slot number, or a symbol number.

[0126] In some embodiments, as shown in Figure 7, the method for performance monitoring of AI / ML models includes the following steps:

[0127] 1. After the performance monitoring of the AI / ML model is triggered, the second device sends the first reference signal of M1 ports and the second reference signal of M2 ports at the same time or in the same time unit.

[0128] 2. The first device obtains the first channel information through an AI / ML model based on the measurement results of the first reference signals at M1 ports;

[0129] 3. The first device measures the second reference signal of M2 ports to obtain the second channel information;

[0130] 4. The first device extracts the channel information corresponding to M2 ports from the first channel information, compares it with the second channel information, and determines whether the AI / ML model is accurate / effective based on the comparison result.

[0131] In case 2, the method for determining the M1 ports and M2 ports can be the same as in case 1, and will not be repeated here.

[0132] In some embodiments, the first reference signal and the second reference signal belong to the same set of reference signals, and the first reference signal and the second reference signal have different indices. For example, the reference signal with the smaller index is the first reference signal, and the reference signal with the larger index is the second reference signal.

[0133] In some embodiments, the first reference signal and the second reference signal belong to different sets of reference signals, and the set of reference signals to which the first reference signal belongs is associated with the set of reference signals to which the second reference signal belongs.

[0134] In this case, since the first reference signal and the second reference signal are sent at the same time / time unit, there is no problem of the first channel information output by the AI / ML model being outdated.

[0135] Case 3

[0136] In some embodiments, M1 ports and M2 ports are two different sets of ports among M3 ports, the first reference signal and the second reference signal are the same reference signal sent at the same time or time unit, and M3 is a positive integer less than or equal to N.

[0137] To avoid the problem of the first channel information output by the AI / ML model being outdated in Case 1, in Case 3, after the performance monitoring of the AI / ML model is triggered, the second device sends reference signals for M3 ports. The measurement results corresponding to the reference signals of a portion of the ports (such as M1 ports) are used as input to the AI / ML model to obtain the first channel information, and the measurement results corresponding to the reference signals of the other portion of the ports (such as M2 ports) are used as the second channel information for performance monitoring.

[0138] In some embodiments, as shown in Figure 8, the method for performance monitoring of AI / ML models includes the following steps:

[0139] 1. After the performance monitoring of the AI / ML model is triggered, the second device sends reference signals to M3 ports;

[0140] 2. The first device obtains the first channel information through an AI / ML model based on the measurement results of the reference signals of M1 out of the M3 ports;

[0141] 3. The first device obtains the second channel information based on the measurement results of the reference signals of M2 out of the M3 ports;

[0142] 4. The first device extracts the channel information corresponding to M2 ports from the first channel information, compares it with the second channel information, and determines whether the AI / ML model is accurate / effective based on the comparison result.

[0143] In case 3, the M3 ports can be obtained in any of the following ways.

[0144] In some embodiments, M3 ports are obtained by beamforming N ports.

[0145] In some embodiments, M3 ports are obtained by sampling N ports in the horizontal and / or vertical dimensions.

[0146] In some embodiments, the M3 ports are the first M3 ports out of the N ports.

[0147] In some embodiments, M3 ports are the last M3 ports out of N ports.

[0148] In some embodiments, M3 ports are obtained by uniformly sampling N ports.

[0149] In some embodiments, the M3 ports are sampled from n port groups obtained by dividing the N ports, where n is a positive integer.

[0150] The method described above for determining M3 ports from N ports is similar to the method described above for determining M1 or M2 ports from N ports. For details, please refer to the description above, which will not be repeated here.

[0151] In some embodiments, M3 is the sum of M1 and M2.

[0152] Optionally, M1 ports are the first M1 ports out of M3 ports, and M2 ports are the last M2 ports out of M3 ports.

[0153] Optionally, M1 ports are the last M1 ports among M3 ports, and M2 ports are the first M2 ports among M3 ports.

[0154] Optionally, M1 ports are the ports with even indices among the M3 ports, and M2 ports are the ports with odd indices among the M3 ports.

[0155] Optionally, M1 ports are the ports with odd indices among the M3 ports, and M2 ports are the ports with even indices among the M3 ports.

[0156] Using the above method, two groups of ports can be further identified from the M3 ports: one group of ports (i.e., M1 ports) is used for model input, and the other group of ports (i.e., M2 ports) is used for model performance monitoring.

[0157] In this case, the problem of the first channel information output by the AI / ML model being outdated, as in Case 1, can also be avoided.

[0158] In some embodiments, in order to improve the accuracy of AI / ML model prediction of first channel information, the input of AI / ML model includes not only third channel information (the third channel information is the channel information corresponding to the M1 ports obtained by measuring the first reference signal of the M1 ports) but also fourth channel information, wherein the fourth channel information is the channel information corresponding to the M4 ports obtained by measuring the third reference signal of the M4 ports.

[0159] As shown in Figure 9, step 410 above can be replaced by step 412 as follows:

[0160] Step 412: The first device obtains the first channel information based on the first reference signals of M1 ports and the third reference signals of M4 ports; wherein the first reference signals are transmitted on the first frequency domain resources, the second reference signals are transmitted on the second frequency domain resources, the third reference signals are transmitted on the third frequency domain resources, and M4 is a positive integer less than or equal to N.

[0161] In some embodiments, the third reference signal is sent by the second device, which sends the third reference signal across M4 ports. The third reference signal and the first reference signal can be the same type of reference signal or two different reference signals; this application does not limit this. Similarly, the third reference signal and the second reference signal can be the same type of reference signal or two different reference signals; this application does not limit this.

[0162] In some embodiments, the first frequency domain resources and the third frequency domain resources correspond to different bandwidths; or, the first frequency domain resources and the third frequency domain resources contain different numbers of subbands; or, the first frequency domain resources and the third frequency domain resources contain different numbers of subcarriers; or, the first frequency domain resources and the third frequency domain resources contain different numbers of PRBs (Physical Resource Blocks).

[0163] In some embodiments, the bandwidth corresponding to the first frequency domain resource is greater than the bandwidth corresponding to the third frequency domain resource; and / or, the bandwidth corresponding to the first frequency domain resource is greater than the bandwidth corresponding to the second frequency domain resource. For example, the first reference signal of M1 ports is transmitted over the full bandwidth, and the second reference signal of M2 ports and the third reference signal of M4 ports are transmitted over a portion of the bandwidth.

[0164] In some embodiments, the subbands included in the second frequency domain resources are different from the subbands included in the third frequency domain resources; or, the subcarriers included in the second frequency domain resources are different from the subcarriers included in the third frequency domain resources; or, the PRBs included in the second frequency domain resources are different from the PRBs included in the third frequency domain resources.

[0165] In some embodiments, M4 ports are N ports, that is, M4 ports are all ports, so that information from all ports can be collected, improving the accuracy of the first channel information output by the AI / ML model.

[0166] In some embodiments, the M4 ports are a subset of the N ports whose indices differ from the M1 ports. This allows for the collection of information from only the M1 ports and also improves the accuracy of the first channel information output by the AI / ML model.

[0167] For example, as shown in Figure 10, the second device transmits a first reference signal with M1 ports (partial ports) on a first frequency domain resource (such as a full-frequency sub-band), as shown by the shaded area filled with diagonal lines on the left side of Figure 10, and transmits a third reference signal with M4 ports (such as full ports) on a third frequency domain resource (such as a partial frequency domain sub-band), as shown by the shaded area filled with diagonal lines on the right side of Figure 10. The first device uses the measurement results of the above two parts of the reference signal as input to the AI / ML model, and the AI / ML model predicts and outputs the first channel information, which includes the channel information corresponding to all N ports on the first frequency domain resource (such as a full-frequency sub-band). This method also considers the frequency domain resource of the reference signal, and uses the reference signal of full ports on the sparse frequency domain sub-band as input to the AI / ML model, which can help obtain the channel information of the full-frequency sub-band, reduce the reference signal overhead, and improve the accuracy of the channel information estimation of the full-frequency sub-band.

[0168] When the input to the AI / ML model includes third-channel information and fourth-channel information, corresponding to cases 1 to 3 above, there are cases 4 to 6 as follows.

[0169] Case 4, as shown in Figure 11, involves the following steps for performance monitoring of AI / ML models:

[0170] 1. At time t1, the second device transmits the first reference signal of M1 ports on the first frequency domain resource and the third reference signal of M4 ports on the third frequency domain resource;

[0171] 2. The first device obtains the first channel information through an AI / ML model based on the measurement results of the first reference signals of M1 ports and the measurement results of the third reference signals of M4 ports; wherein, the first channel information includes the channel information corresponding to N ports in the first frequency domain resources;

[0172] 3. After the performance monitoring of the AI / ML model is triggered, the second device sends the second reference signal of M2 ports on the second frequency domain resources at time t2;

[0173] 4. The first device measures the second reference signal of M2 ports to obtain the second channel information; wherein, the second channel information includes the channel information corresponding to the M2 ports in the second frequency domain resource;

[0174] 5. The first device extracts the channel information corresponding to the M2 ports in the second frequency domain resources from the first channel information, compares it with the second channel information, and determines whether the AI / ML model is accurate / effective based on the comparison result.

[0175] Case 5, as shown in Figure 12, involves the following steps for performance monitoring of AI / ML models:

[0176] 1. After the performance monitoring of the AI / ML model is triggered, the second device sends the first reference signal of M1 ports on the first frequency domain resource, the second reference signal of M2 ports on the second frequency domain resource, and the third reference signal of M4 ports on the third frequency domain resource at the same time or in the same time unit.

[0177] 2. The first device obtains the first channel information through an AI / ML model based on the measurement results of the first reference signals of M1 ports and the measurement results of the third reference signals of M4 ports; wherein, the first channel information includes the channel information corresponding to N ports in the first frequency domain resources;

[0178] 3. The first device measures the second reference signal of M2 ports to obtain the second channel information; wherein, the second channel information includes the channel information corresponding to the M2 ports in the second frequency domain resource;

[0179] 4. The first device extracts the channel information corresponding to M2 ports on the second frequency domain resources from the first channel information, compares it with the second channel information, and determines whether the AI / ML model is accurate / effective based on the comparison result.

[0180] In some embodiments, in case 5, the first reference signal, the second reference signal, and the third reference signal belong to the same set of reference signals, and the first reference signal, the second reference signal, and the third reference signal have different indices. For example, the reference signals with smaller indices are the first reference signal and the third reference signal, used as input to the model, while the reference signal with larger indices is the second reference signal, used for model performance monitoring.

[0181] In some embodiments, in Case 5, the first frequency domain resource is the full bandwidth, while the second and third frequency domain resources are partial bandwidths. The number of subbands / subcarriers / PRBs included in the second and third frequency domain resources may be the same or different.

[0182] For example, the first reference signal is the CSI-RS of M1 ports on the first frequency domain resource, which includes 18 sub-bands, M1 = 16; the third reference signal is the CSI-RS of M4 ports on the third frequency domain resource, which includes 4 sub-bands, M4 = N = 128; the second reference signal is the CSI-RS of M2 ports on the second frequency domain resource, which includes 4 sub-bands, and the 4 sub-bands have the same or different sub-band indices as the third frequency domain resource, M2 < N; where N is the total number of ports, i.e., the maximum number of configurable ports.

[0183] Case 6, as shown in Figure 13, involves the following steps for performance monitoring of AI / ML models:

[0184] 1. After the performance monitoring of the AI / ML model is triggered, the second device sends the first reference signal of M1 ports on the first frequency domain resource at the same time or in the same time unit, and sends the third reference signal of M4 ports on the fourth frequency domain resource; wherein, the fourth frequency domain resource includes the second frequency domain resource and the third frequency domain resource;

[0185] 2. The first device obtains the first channel information through an AI / ML model based on the measurement results of the first reference signals of M1 ports and the measurement results of the third reference signals of M4 ports transmitted on the third frequency domain resource; wherein, the first channel information includes the channel information corresponding to N ports on the first frequency domain resource;

[0186] 3. The first device obtains the second channel information based on the measurement results of the third reference signals transmitted on the M4 ports in the second frequency domain resource; wherein, the second channel information includes the channel information corresponding to the M4 ports in the second frequency domain resource;

[0187] 4. The first device extracts the channel information corresponding to the M4 ports in the second frequency domain resources from the first channel information, compares it with the second channel information, and determines whether the AI / ML model is accurate / effective based on the comparison result.

[0188] For example, the first reference signal is the CSI-RS of M1 ports on a first frequency domain resource, which includes 18 sub-bands, M1 = 16; the third reference signal is the CSI-RS of M4 ports on a fourth frequency domain resource, which includes 8 sub-bands, M4 = N = 128. Optionally, the third frequency domain resource is the first 4 sub-bands of the fourth frequency domain resource, used as input to the AI / ML model, and the second frequency domain resource is the last 4 sub-bands of the fourth frequency domain resource, used for model performance monitoring.

[0189] In the above manner, the first device obtains the first channel information based on the first reference signals of M1 ports and the third reference signals of M4 ports. Combining the reference signal measurement results of some ports across the full bandwidth and the reference signal measurement results of all ports across a portion of the bandwidth, the first channel information is predicted using an AI / ML model. This helps improve the accuracy of the first channel information output by the AI / ML model, thereby enabling an accurate evaluation of the performance of the AI / ML model.

[0190] In the above embodiment, when M2 is less than N, the reference signal overhead for performance monitoring of the AI / ML model can be reduced. However, it requires separating the channel information corresponding to M2 ports from the first channel information output by the AI / ML model. In some cases, it may be impossible to separate the channel information corresponding to M2 ports, for example, if the output of the AI / ML model is a feature vector. Therefore, an alternative solution is to use a full-port reference signal, i.e., M2 equals N, when monitoring the performance of the AI / ML model. However, this results in a larger reference signal overhead. When M2 equals N, the transmission overhead of the second reference signal is negatively correlated with the interval between measurements of the second channel information. By setting the trigger interval for performance monitoring of the AI / ML model to a longer time, i.e., a longer interval between measurements of the second channel information, the reference signal overhead can be reduced from a time domain perspective.

[0191] In some embodiments, the verification is triggered in at least one of the following ways: by the second device through explicit or implicit means; or by the first device through a request. Optionally, the performance monitoring of the AI / ML model can be triggered by the second device. For example, the second device can trigger it explicitly through signaling, or implicitly through periodic, non-periodic, or semi-persistent transmission of reference signals. Optionally, the performance monitoring of the AI / ML model can also be triggered by the first device. For example, the first device can send a request message to the second device requesting performance monitoring of the AI / ML model, and after receiving the request message, the second device can send a reference signal to the first device for performance monitoring of the AI / ML model.

[0192] In some embodiments, if the verification result of the first channel information is inaccurate / invalid, the method provided in this application further includes at least one of the following: updating the parameters of the AI / ML model that outputs the first channel information; switching to the reference signal measurement results based on N ports to obtain the channel information corresponding to the N ports; and switching the AI / ML model that outputs the first channel information.

[0193] For example, when the first device is a terminal device and the second device is a network device, the process for updating the parameters of the AI / ML model that outputs the first channel information may include the following steps: the first device sends a model update request to the second device, which requests an update to the parameters of the AI / ML model; the second device updates the parameters of the AI / ML model according to the model update request and sends the updated model parameters to the first device.

[0194] For example, switching to a reference signal measurement based on N ports to obtain channel information corresponding to those N ports is equivalent to switching to a traditional CSI measurement and feedback scheme. For example, when the first device is a terminal device and the second device is a network device, the process may include the following: the first device sends a first handover request to the second device, which requests a switch to a traditional CSI measurement and feedback scheme; the second device sends reference signals from the N ports according to the first handover request.

[0195] For example, when the first device is a terminal device and the second device is a network device, the process for switching the AI / ML model that outputs the first channel information may include the following steps: the first device sends a second switching request to the second device, the second switching request being used to request the switching of the AI / ML model; the second device determines the AI / ML model after switching based on the second switching request and instructs the first device on the switched AI / ML model.

[0196] For example, if the first device is a terminal device and the second device is a network device, and the verification result of the first channel information is inaccurate / invalid, the first device may also send recommended reference signal configuration information and / or recommended model configuration information to the second device. The recommended reference signal configuration information indicates a recommended reference signal configuration, and may include at least one of the following: the port for transmitting the reference signal, and the frequency domain resources for transmitting the reference signal. The recommended model configuration information indicates a recommended model configuration, and may include at least one of the following: recommended model parameters, and a recommended model index.

[0197] For example, if the first device is a terminal device and the second device is a network device, and the verification result of the first channel information is inaccurate / invalid, the first device can also send feedback information to the second device. This feedback information indicates that the verification result of the first channel information is inaccurate / invalid, or indicates that the AI / ML model currently used by the first device is inaccurate / invalid. After receiving this feedback information, the second device decides on the subsequent processing method. For example, it may update the parameters of the AI / ML model that outputs the first channel information, or switch to a traditional CSI measurement and feedback scheme, or switch the AI / ML model that outputs the first channel information, or adjust the CSI measurement and feedback scheme according to the recommended reference signal configuration information and / or the recommended model configuration information.

[0198] In some embodiments, the first channel information is output by an AI / ML model, and the first channel information is not outdated if any of the following conditions are met:

[0199] (1) The time interval between the AI / ML model outputting the first channel information and the time verifying the first channel information is less than or equal to the first interval.

[0200] (2) The time when the AI / ML model outputs the first channel information and the time when the first channel information is verified are within the same time window;

[0201] (3) The first channel information has not exceeded the validity period.

[0202] In the first approach, a first interval duration can be pre-configured or defined. If the interval between the moment the AI / ML model outputs the first channel information and the moment the first channel information is verified is less than or equal to this first interval duration, it indicates that the first channel information is not outdated. In this case, the method described in scenario 1 or scenario 4 above can be applied to monitor the performance of the AI / ML model. The first interval duration can be configured by the network device, pre-configured, or agreed upon by the protocol.

[0203] In the second approach, a time window can be pre-configured or defined, encompassing the moment the AI / ML model outputs the first channel information and the moment the first channel information is verified. If the information falls within this time window, it indicates that the first channel information is not outdated, and the methods described in scenario 1 or 4 above can be applied to monitor the performance of the AI / ML model. The time window can be configured by the network device, pre-configured, or agreed upon by the protocol.

[0204] In the third approach, a valid time period can be pre-configured or defined. If the first channel information does not exceed the valid time period, it indicates that the first channel information is not expired, and the methods described in cases 1 or 4 above can be applied to monitor the performance of the AI / ML model. For example, the valid time period can be 10 milliseconds, starting from the moment the AI / ML model outputs the first channel information. Within this valid time period, the first channel information is not expired; after this valid time period, the first channel information expires. The valid time period can be configured by the network device, pre-configured, or agreed upon by the protocol.

[0205] Furthermore, the interval duration, time window duration, or effective time in the above methods can be expressed as a value calculated in time units (e.g., milliseconds, microseconds, etc.); or it can be expressed as an SFN value, time slot number, or symbol number. The terminal device can first report the suggested values, and then the network device can configure them.

[0206] In some embodiments, the method used to obtain the first channel information and / or the second channel information is configured by the network device itself, or, if the first device is a terminal device, it can be configured by the network device according to the capabilities of the first device. That is, the method used for performance monitoring of the AI / ML model in any of the above scenarios 1-6 can be configured by the network device itself, or it can be configured by the network device according to the capabilities of the terminal device. In the case where the network device configures the method itself, the terminal device uses the method configured by the network device to perform performance monitoring of the AI / ML model. In the case where the network device configures the method according to the capabilities of the terminal device, the terminal device can first report to the network device the methods it supports and / or does not support, and the network device then configures the method for performance monitoring of the AI / ML model for the terminal device according to the terminal's capabilities.

[0207] The above embodiments only describe the technical solution provided by this application from the perspective of the interaction between the first device and the second device. The steps performed by the first device described above can be implemented independently as a verification method on the first device side. Similarly, the steps performed by the second device described above can be implemented independently as a verification method on the second device side. Furthermore, the letters M1, M2, M3, M4, etc., used in this application to represent quantities are for ease of description. This application does not limit the use of other letters, symbols, or descriptions such as "first quantity," "second quantity," etc., to distinguish different quantities.

[0208] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0209] Please refer to Figure 14, which shows a block diagram of a verification device provided in one embodiment of this application. This device has the function of implementing the verification method executed by the first device described above. This function can be implemented in hardware or by hardware executing corresponding software. This device can be the first device described above, or it can be disposed within the first device. As shown in Figure 14, the device 1400 may include: a processing module 1410.

[0210] The processing module 1410 is used to obtain first channel information based on the first reference signals of M1 ports. The first channel information is the channel information corresponding to N ports, where N is an integer greater than 1 and M1 is a positive integer less than N.

[0211] The processing module 1410 is further configured to obtain second channel information based on the second reference signals of the M2 ports, wherein the second channel information is the channel information corresponding to the M2 ports, and M2 is a positive integer less than or equal to N.

[0212] The processing module 1410 is further configured to verify the first channel information based on the second channel information.

[0213] In some embodiments, the first channel information is output by an AI / ML model, and the input of the AI / ML model includes third channel information, which is the channel information corresponding to the M1 ports obtained by measuring the first reference signal of the M1 ports.

[0214] In some embodiments, the first reference signal and the second reference signal are transmitted at different times.

[0215] In some embodiments, the first reference signal and the second reference signal are transmitted at the same time; or, the first reference signal and the second reference signal are transmitted in the same time unit.

[0216] In some embodiments, the first reference signal and the second reference signal belong to the same set of reference signals, and the first reference signal and the second reference signal have different indices; or, the first reference signal and the second reference signal belong to different sets of reference signals, and the set of reference signals to which the first reference signal belongs is associated with the set of reference signals to which the second reference signal belongs.

[0217] In some embodiments, the M1 ports and the M2 ports are completely identical; or, the M1 ports and the M2 ports are completely different; or, the M1 ports and the M2 ports are partially identical.

[0218] In some embodiments, the M1 ports and / or the M2 ports are obtained by beamforming the N ports; or, the M1 ports and / or the M2 ports are obtained by sampling the N ports in the horizontal and / or vertical dimensions; or, the M1 ports are the first M1 ports among the N ports, and / or, the M2 ports are the first M2 ports among the N ports; or, the M1 ports are the last M1 ports among the N ports, and / or, the M2 ports are the last M2 ports among the N ports; or, the M1 ports and / or the M2 ports are obtained by uniformly sampling the N ports; or, the M1 ports and / or the M2 ports are obtained by sampling from n port groups divided from the N ports, where n is a positive integer.

[0219] In some embodiments, the M1 ports and the M2 ports are two different sets of ports among the M3 ports, the first reference signal and the second reference signal are the same reference signal sent at the same time or time unit, and M3 is a positive integer less than or equal to N.

[0220] In some embodiments, M3 is the sum of M1 and M2; the M1 ports are the first M1 ports among the M3 ports, and the M2 ports are the last M2 ports among the M3 ports; or, the M1 ports are the last M1 ports among the M3 ports, and the M2 ports are the first M2 ports among the M3 ports; or, the M1 ports are the ports with even-numbered indices among the M3 ports, and the M2 ports are the ports with odd-numbered indices among the M3 ports; or, the M1 ports are the ports with odd-numbered indices among the M3 ports, and the M2 ports are the ports with even-numbered indices among the M3 ports.

[0221] In some embodiments, the M3 ports are obtained by beamforming the N ports; or, the M3 ports are obtained by sampling the N ports in the horizontal and / or vertical dimensions; or, the M3 ports are the first M3 ports among the N ports; or, the M3 ports are the last M3 ports among the N ports; or, the M3 ports are obtained by uniformly sampling the N ports; or, the M3 ports are obtained by sampling from n port groups divided from the N ports, where n is a positive integer.

[0222] In some embodiments, the processing module 1410 is configured to obtain the first channel information based on the first reference signals of the M1 ports and the third reference signals of the M4 ports; wherein the first reference signals are transmitted on the first frequency domain resources, the second reference signals are transmitted on the second frequency domain resources, the third reference signals are transmitted on the third frequency domain resources, and M4 is a positive integer less than or equal to N.

[0223] In some embodiments, the first frequency domain resource and the third frequency domain resource correspond to different bandwidths; or, the first frequency domain resource and the third frequency domain resource contain different numbers of subbands; or, the first frequency domain resource and the third frequency domain resource contain different numbers of subcarriers; or, the first frequency domain resource and the third frequency domain resource contain different numbers of PRBs.

[0224] In some embodiments, the bandwidth corresponding to the first frequency domain resource is greater than the bandwidth corresponding to the third frequency domain resource; and / or, the bandwidth corresponding to the first frequency domain resource is greater than the bandwidth corresponding to the second frequency domain resource.

[0225] In some embodiments, the subbands included in the second frequency domain resource are different from the subbands included in the third frequency domain resource; or, the subcarriers included in the second frequency domain resource are different from the subcarriers included in the third frequency domain resource; or, the PRBs included in the second frequency domain resource are different from the PRBs included in the third frequency domain resource.

[0226] In some embodiments, the M4 ports are the N ports, or the M4 ports are a subset of the N ports whose indices differ from the M1 ports.

[0227] In some embodiments, the processing module 1410 is configured to compare the second channel information with the channel information of the corresponding M2 ports in the first channel information to obtain a comparison result; and to verify the first channel information based on the comparison result.

[0228] In some embodiments, the comparison result is used to characterize the SGCS between the second channel information and the channel information of the corresponding M2 ports in the first channel information; or, the comparison result is used to characterize the difference between the second channel information and the channel information of the corresponding M2 ports in the first channel information; or, the comparison result is used to characterize the correlation between the second channel information and the channel information of the corresponding M2 ports in the first channel information.

[0229] In some embodiments, when M2 equals N, the transmission overhead of the second reference signal is negatively correlated with the interval duration for measuring the second channel information.

[0230] In some embodiments, the verification is triggered in at least one of the following ways: by the second device in an explicit or implicit manner; or by the first device through a request.

[0231] In some embodiments, if the verification result of the first channel information is inaccurate or invalid, the method further includes at least one of the following: updating the parameters of the AI / ML model that outputs the first channel information; switching to the reference signal measurement results based on the N ports to obtain the channel information corresponding to the N ports; and switching the AI / ML model that outputs the first channel information.

[0232] In some embodiments, the first channel information is output by an AI / ML model, and the first channel information is not outdated if the following conditions are met: the time interval between the time when the AI / ML model outputs the first channel information and the time when the first channel information is verified is less than or equal to a first interval; or, the time when the AI / ML model outputs the first channel information and the time when the first channel information is verified are within the same time window; or, the first channel information has not exceeded its validity period.

[0233] In some embodiments, the method used to obtain the first channel information and / or the second channel information is configured by the network device itself, or by the network device according to the capabilities of the first device.

[0234] Please refer to Figure 15, which shows a block diagram of a verification device provided in another embodiment of this application. This device has the function of implementing the verification method executed by the second device described above. This function can be implemented in hardware or by hardware executing corresponding software. This device can be the second device described above, or it can be disposed within the second device. As shown in Figure 15, the device 1500 may include: a sending module 1510.

[0235] The transmitting module 1510 is used to transmit first reference signals for M1 ports. The first reference signals for M1 ports are used to determine first channel information. The first channel information is channel information corresponding to N ports, where N is an integer greater than 1 and M1 is a positive integer less than N.

[0236] The transmitting module is further configured to transmit second reference signals for M2 ports, the second reference signals for M2 ports being used to determine second channel information, the second channel information being the channel information corresponding to the M2 ports, where M2 is a positive integer less than or equal to N.

[0237] The second channel information is used to verify the first channel information.

[0238] In some embodiments, the first channel information is output by an AI / ML model, and the input of the AI / ML model includes third channel information, which is the channel information corresponding to the M1 ports obtained by measuring the first reference signal of the M1 ports.

[0239] In some embodiments, the first reference signal and the second reference signal are transmitted at different times.

[0240] In some embodiments, the first reference signal and the second reference signal are transmitted at the same time; or, the first reference signal and the second reference signal are transmitted in the same time unit.

[0241] In some embodiments, the first reference signal and the second reference signal belong to the same set of reference signals, and the first reference signal and the second reference signal have different indices; or, the first reference signal and the second reference signal belong to different sets of reference signals, and the set of reference signals to which the first reference signal belongs is associated with the set of reference signals to which the second reference signal belongs.

[0242] In some embodiments, the M1 ports and the M2 ports are completely identical; or, the M1 ports and the M2 ports are completely different; or, the M1 ports and the M2 ports are partially identical.

[0243] In some embodiments, the M1 ports and / or the M2 ports are obtained by beamforming the N ports; or, the M1 ports and / or the M2 ports are obtained by sampling the N ports in the horizontal and / or vertical dimensions; or, the M1 ports are the first M1 ports among the N ports, and / or, the M2 ports are the first M2 ports among the N ports; or, the M1 ports are the last M1 ports among the N ports, and / or, the M2 ports are the last M2 ports among the N ports; or, the M1 ports and / or the M2 ports are obtained by uniformly sampling the N ports; or, the M1 ports and / or the M2 ports are obtained by sampling from n port groups divided from the N ports, where n is a positive integer.

[0244] In some embodiments, the M1 ports and the M2 ports are two different sets of ports among the M3 ports, the first reference signal and the second reference signal are the same reference signal sent at the same time or time unit, and M3 is a positive integer less than or equal to N.

[0245] In some embodiments, M3 is the sum of M1 and M2; the M1 ports are the first M1 ports among the M3 ports, and the M2 ports are the last M2 ports among the M3 ports; or, the M1 ports are the last M1 ports among the M3 ports, and the M2 ports are the first M2 ports among the M3 ports; or, the M1 ports are the ports with even-numbered indices among the M3 ports, and the M2 ports are the ports with odd-numbered indices among the M3 ports; or, the M1 ports are the ports with odd-numbered indices among the M3 ports, and the M2 ports are the ports with even-numbered indices among the M3 ports.

[0246] In some embodiments, the M3 ports are obtained by beamforming the N ports; or, the M3 ports are obtained by sampling the N ports in the horizontal and / or vertical dimensions; or, the M3 ports are the first M3 ports among the N ports; or, the M3 ports are the last M3 ports among the N ports; or, the M3 ports are obtained by uniformly sampling the N ports; or, the M3 ports are obtained by sampling from n port groups divided from the N ports, where n is a positive integer.

[0247] In some embodiments, the transmitting module 1510 is further configured to transmit a third reference signal for M4 ports, wherein the third reference signal for M4 ports is used to obtain the first channel information by combining it with the first reference signal for M1 ports; wherein the first reference signal is transmitted on a first frequency domain resource, the second reference signal is transmitted on a second frequency domain resource, the third reference signal is transmitted on a third frequency domain resource, and M4 is a positive integer less than or equal to N.

[0248] In some embodiments, the first frequency domain resource and the third frequency domain resource correspond to different bandwidths; or, the first frequency domain resource and the third frequency domain resource contain different numbers of subbands; or, the first frequency domain resource and the third frequency domain resource contain different numbers of subcarriers; or, the first frequency domain resource and the third frequency domain resource contain different numbers of PRBs.

[0249] In some embodiments, the bandwidth corresponding to the first frequency domain resource is greater than the bandwidth corresponding to the third frequency domain resource; and / or, the bandwidth corresponding to the first frequency domain resource is greater than the bandwidth corresponding to the second frequency domain resource.

[0250] In some embodiments, the subbands included in the second frequency domain resource are different from the subbands included in the third frequency domain resource; or, the subcarriers included in the second frequency domain resource are different from the subcarriers included in the third frequency domain resource; or, the PRBs included in the second frequency domain resource are different from the PRBs included in the third frequency domain resource.

[0251] In some embodiments, the M4 ports are the N ports, or the M4 ports are a subset of the N ports whose indices differ from the M1 ports.

[0252] In some embodiments, the second channel information is compared with the channel information of the corresponding M2 ports in the first channel information, and the comparison result is used to verify the first channel information.

[0253] In some embodiments, the comparison result is used to characterize the SGCS between the second channel information and the channel information of the corresponding M2 ports in the first channel information; or, the comparison result is used to characterize the difference between the second channel information and the channel information of the corresponding M2 ports in the first channel information; or, the comparison result is used to characterize the correlation between the second channel information and the channel information of the corresponding M2 ports in the first channel information.

[0254] In some embodiments, when M2 equals N, the transmission overhead of the second reference signal is negatively correlated with the interval duration for measuring the second channel information.

[0255] In some embodiments, the verification is triggered in at least one of the following ways: by the second device in an explicit or implicit manner; or by the first device through a request.

[0256] In some embodiments, if the verification result of the first channel information is inaccurate or invalid, the method further includes at least one of the following: updating the parameters of the AI / ML model that outputs the first channel information; switching to the reference signal measurement results based on the N ports to obtain the channel information corresponding to the N ports; and switching the AI / ML model that outputs the first channel information.

[0257] In some embodiments, the first channel information is output by an AI / ML model, and the first channel information is not outdated if the following conditions are met: the time interval between the time when the AI / ML model outputs the first channel information and the time when the first channel information is verified is less than or equal to a first interval; or, the time when the AI / ML model outputs the first channel information and the time when the first channel information is verified are within the same time window; or, the first channel information has not exceeded its validity period.

[0258] In some embodiments, the method used to obtain the first channel information and / or the second channel information is configured by the network device itself, or by the network device according to the capabilities of the first device.

[0259] It should be noted that the above embodiments only illustrate the division of the above functional modules when implementing the device. In actual applications, the above functions can be assigned to different functional modules according to actual needs, that is, the content structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0260] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here. For details not described in detail in the apparatus embodiments, please refer to the above method embodiments.

[0261] Please refer to Figure 16, which shows a schematic diagram of a communication device according to an embodiment of this application. The communication device 1600 may include a processor 1601, a transceiver 1602, and a memory 1603. The transceiver 1602 is used to implement sending and receiving functions, such as the functions of the sending and receiving modules described above. The processor can be used to implement other processing functions or control sending and / or receiving, such as the functions of the processing modules described above.

[0262] The processor 1601 includes one or more processing cores, and the processor 1601 executes various functional applications and information processing by running software programs and modules.

[0263] The transceiver 1602 may include a receiver and a transmitter, for example, the receiver and transmitter may be implemented as the same wireless communication component, which may include a wireless communication chip and a radio frequency antenna.

[0264] The memory 1603 can be connected to the processor 1601 and the transceiver 1602.

[0265] The memory 1603 can be used to store computer programs executed by the processor, and the processor 1601 is used to execute the computer program.

[0266] In some embodiments, when the communication device is a first device, the processor 1601 is configured to obtain first channel information based on first reference signals of M1 ports, wherein the first channel information is channel information corresponding to N ports, where N is an integer greater than 1 and M1 is a positive integer less than N. The processor 1601 is further configured to obtain second channel information based on second reference signals of M2 ports, wherein the second channel information is channel information corresponding to the M2 ports, where M2 is a positive integer less than or equal to N. The processor 1601 is further configured to verify the first channel information based on the second channel information.

[0267] In some embodiments, when the communication device is a second device, the transceiver 1602 is used to transmit first reference signals for M1 ports. The first reference signals for the M1 ports are used to determine first channel information, which is channel information corresponding to N ports, where N is an integer greater than 1, and M1 is a positive integer less than N. The transceiver 1602 is also used to transmit second reference signals for M2 ports. The second reference signals for the M2 ports are used to determine second channel information, which is channel information corresponding to the M2 ports, where M2 is a positive integer less than or equal to N. The second channel information is used to verify the first channel information.

[0268] For details not described in this embodiment, please refer to the embodiments above, which will not be repeated here.

[0269] Furthermore, the memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, including but not limited to: magnetic disks or optical disks, electrically erasable programmable read-only memory, erasable programmable read-only memory, statically accessible memory, read-only memory, magnetic memory, flash memory, and programmable read-only memory.

[0270] This application also provides a computer-readable storage medium storing a computer program for execution by a processor to implement the verification method executed by the first device or the verification method executed by the second device. In some embodiments, the computer-readable storage medium may include ROM (Read-Only Memory), RAM (Random-Access Memory), SSD (Solid State Drives), or optical disc, etc. The random access memory may include ReRAM (Resistance Random Access Memory) and DRAM (Dynamic Random Access Memory).

[0271] This application also provides a chip, which includes programmable logic circuits and / or program instructions. When the chip is running, it is used to implement the verification method executed by the first device or the verification method executed by the second device.

[0272] This application also provides a computer program product, which includes computer instructions stored in a computer-readable storage medium. A processor reads and executes the computer instructions from the computer-readable storage medium to implement the verification method executed by the first device or the verification method executed by the second device.

[0273] It should be understood that the term "instruction" mentioned in the embodiments of this application can be a direct instruction, an indirect instruction, or an indication of a relationship. For example, A instructing B can mean that A directly instructs B, such as B being able to obtain information through A; it can also mean that A indirectly instructs B, such as A instructing C, so B can obtain information through C; or it can mean that there is a relationship between A and B.

[0274] In the description of the embodiments of this application, the term "correspondence" may indicate that there is a direct or indirect correspondence between two things, or that there is an association between two things, or that there is a relationship of instruction and being instructed, configuration and being configured, etc.

[0275] In some embodiments of this application, "predefined" can be achieved by pre-storing corresponding codes, tables, or other means that can be used to indicate relevant information in the device (e.g., including terminal devices and network devices). This application does not limit the specific implementation method. For example, predefined can refer to what is defined in the protocol.

[0276] In some embodiments of this application, the term "protocol" may refer to standard protocols in the field of communications, such as LTE protocols, NR protocols, and related protocols applied in future communication systems. This application does not limit the scope of these protocols.

[0277] In this article, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0278] In this article, "greater than or equal to" can mean greater than or equal to, and "less than or equal to" can mean less than or equal to.

[0279] Furthermore, the step numbers described herein are merely illustrative of one possible execution order between steps. In some other embodiments, the steps may not be executed in the order of their numbers, such as two steps with different numbers being executed simultaneously, or two steps with different numbers being executed in the reverse order of the illustration. This application does not limit this.

[0280] Those skilled in the art will recognize that the functions described in the embodiments of this application in one or more of the above examples can be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transfer of a computer program from one place to another. Storage media can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0281] The above description is merely an exemplary embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A verification method, characterized in that, The method is performed by a first device, and the method includes: Based on the first reference signals of M1 ports, the first channel information is obtained. The first channel information is the channel information corresponding to N ports, where N is an integer greater than 1 and M1 is a positive integer less than N. Based on the second reference signals of the M2 ports, the second channel information is obtained. The second channel information is the channel information corresponding to the M2 ports, where M2 is a positive integer less than or equal to N. The first channel information is verified based on the second channel information.

2. The method according to claim 1, characterized in that, The first channel information is output by an artificial intelligence (AI) / machine learning (ML) model. The input of the AI / ML model includes third channel information, which is the channel information corresponding to the M1 ports obtained by measuring the first reference signal of the M1 ports.

3. The method according to claim 1 or 2, characterized in that, The first reference signal and the second reference signal are transmitted at different times.

4. The method according to claim 1 or 2, characterized in that, The first reference signal and the second reference signal are transmitted at the same time; or, the first reference signal and the second reference signal are transmitted in the same time unit.

5. The method according to claim 4, characterized in that, The first reference signal and the second reference signal belong to the same set of reference signals, and the first reference signal and the second reference signal have different indices; or, The first reference signal and the second reference signal belong to different sets of reference signals, and the set of reference signals to which the first reference signal belongs is related to the set of reference signals to which the second reference signal belongs.

6. The method according to any one of claims 3 to 5, characterized in that, The M1 ports and the M2 ports are exactly the same; or, The M1 ports and the M2 ports are completely different; or, The M1 ports and the M2 ports are partially the same.

7. The method according to any one of claims 3 to 6, characterized in that, The M1 ports and / or the M2 ports are obtained by beamforming the N ports; or... The M1 ports and / or the M2 ports are obtained by sampling the N ports in the horizontal and / or vertical dimensions. or, The M1 ports are the first M1 ports out of the N ports, and / or the M2 ports are the first M2 ports out of the N ports; or, The M1 ports are the last M1 ports among the N ports, and / or the M2 ports are the last M2 ports among the N ports; or, The M1 ports and / or the M2 ports are obtained by uniformly sampling the N ports; or, The M1 ports and / or the M2 ports are sampled from the n port groups obtained by dividing the N ports, where n is a positive integer.

8. The method according to claim 1 or 2, characterized in that, The M1 ports and the M2 ports are two different sets of ports among the M3 ports. The first reference signal and the second reference signal are the same reference signal sent at the same time or time unit. M3 is a positive integer less than or equal to N.

9. The method according to claim 8, characterized in that, M3 is the sum of M1 and M2; The M1 ports are the first M1 ports out of the M3 ports, and the M2 ports are the last M2 ports out of the M3 ports; or, The M1 ports are the last M1 ports among the M3 ports, and the M2 ports are the first M2 ports among the M3 ports; or, The M1 ports are the ports with even indices among the M3 ports, and the M2 ports are the ports with odd indices among the M3 ports; or, The M1 ports are the ports with odd-numbered indices among the M3 ports, and the M2 ports are the ports with even-numbered indices among the M3 ports.

10. The method according to claim 8 or 9, characterized in that, The M3 ports are obtained by beamforming the N ports; or... The M3 ports are obtained by sampling the N ports in the horizontal and / or vertical dimensions; or... The M3 ports are the first M3 ports out of the N ports; or, The M3 ports are the last M3 ports among the N ports; or, The M3 ports are obtained by uniformly sampling the N ports; or, The M3 ports are sampled from the n port groups obtained by dividing the N ports, where n is a positive integer.

11. The method according to any one of claims 1 to 10, characterized in that, The step of obtaining the first channel information based on the first reference signals of M1 ports includes: The first channel information is obtained based on the first reference signals of the M1 ports and the third reference signals of the M4 ports; Wherein, the first reference signal is transmitted on the first frequency domain resource, the second reference signal is transmitted on the second frequency domain resource, the third reference signal is transmitted on the third frequency domain resource, and M4 is a positive integer less than or equal to N.

12. The method according to claim 11, characterized in that, The first frequency domain resource and the third frequency domain resource correspond to different bandwidths; or, The first frequency domain resource and the third frequency domain resource contain different numbers of subbands; or, The first frequency domain resource and the third frequency domain resource contain different numbers of subcarriers; or, The first frequency domain resource and the third frequency domain resource contain different numbers of Physical Resource Blocks (PRBs).

13. The method according to claim 11 or 12, characterized in that, The bandwidth corresponding to the first frequency domain resource is greater than the bandwidth corresponding to the third frequency domain resource; and / or, The bandwidth corresponding to the first frequency domain resource is greater than the bandwidth corresponding to the second frequency domain resource.

14. The method according to any one of claims 11 to 13, characterized in that, The subbands included in the second frequency domain resource are different from the subbands included in the third frequency domain resource; or, The subcarriers included in the second frequency domain resource are different from the subcarriers included in the third frequency domain resource; or, The PRBs included in the second frequency domain resource are different from those included in the third frequency domain resource.

15. The method according to any one of claims 11 to 14, characterized in that, The M4 ports are the N ports, or the M4 ports are a subset of the N ports whose indices differ from the M1 ports.

16. The method according to any one of claims 1 to 15, characterized in that, The step of verifying the first channel information based on the second channel information includes: The second channel information is compared with the channel information of the corresponding M2 ports in the first channel information to obtain the comparison result; Based on the comparison results, the first channel information is verified.

17. The method according to claim 16, characterized in that, The comparison result is used to characterize the squared generalized cosine similarity (SGCS) between the second channel information and the channel information of the corresponding M2 ports in the first channel information; or, The comparison result is used to characterize the difference between the second channel information and the channel information of the corresponding M2 ports in the first channel information; or, The comparison result is used to characterize the correlation between the second channel information and the channel information of the corresponding M2 ports in the first channel information.

18. The method according to any one of claims 1 to 17, characterized in that, When M2 equals N, the transmission overhead of the second reference signal is negatively correlated with the interval duration for measuring the second channel information.

19. The method according to any one of claims 1 to 18, characterized in that, The verification is triggered by at least one of the following methods: Triggered by a second device either explicitly or implicitly; It is triggered by the first device through a request.

20. The method according to any one of claims 1 to 19, characterized in that, If the verification result of the first channel information is inaccurate or invalid, the method further includes at least one of the following: Update the parameters of the AI / ML model that outputs the first channel information; Switch to the reference signal measurement results based on the N ports to obtain the channel information corresponding to the N ports; Switch the AI / ML model that outputs the first channel information.

21. The method according to any one of claims 1 to 20, characterized in that, The first channel information is output by the AI / ML model, and it is not outdated if the following conditions are met: The time interval between the moment when the AI / ML model outputs the first channel information and the moment when the first channel information is verified is less than or equal to the first interval duration. or, The time when the AI / ML model outputs the first channel information and the time when the first channel information is verified are within the same time window; or, The first channel information did not exceed the validity period.

22. The method according to any one of claims 1 to 21, characterized in that, The method used to obtain the first channel information and / or the second channel information is configured by the network device itself, or by the network device according to the capabilities of the first device.

23. A verification method, characterized in that, The method is performed by a second device, and the method includes: Send first reference signals for M1 ports, wherein the first reference signals for M1 ports are used to determine first channel information, wherein the first channel information is channel information corresponding to N ports, where N is an integer greater than 1 and M1 is a positive integer less than N; Send second reference signals for M2 ports, the second reference signals for M2 ports are used to determine second channel information, the second channel information is the channel information corresponding to the M2 ports, and M2 is a positive integer less than or equal to N; The second channel information is used to verify the first channel information.

24. The method according to claim 23, characterized in that, The first channel information is output by an artificial intelligence (AI) / machine learning (ML) model. The input of the AI / ML model includes third channel information, which is the channel information corresponding to the M1 ports obtained by measuring the first reference signal of the M1 ports.

25. The method according to claim 23 or 24, characterized in that, The first reference signal and the second reference signal are transmitted at different times.

26. The method according to claim 23 or 24, characterized in that, The first reference signal and the second reference signal are transmitted at the same time; or, the first reference signal and the second reference signal are transmitted in the same time unit.

27. The method according to claim 26, characterized in that, The first reference signal and the second reference signal belong to the same set of reference signals, and the first reference signal and the second reference signal have different indices; or, The first reference signal and the second reference signal belong to different sets of reference signals, and the set of reference signals to which the first reference signal belongs is related to the set of reference signals to which the second reference signal belongs.

28. The method according to any one of claims 25 to 27, characterized in that, The M1 ports and the M2 ports are exactly the same; or, The M1 ports and the M2 ports are completely different; or, The M1 ports and the M2 ports are partially the same.

29. The method according to any one of claims 25 to 28, characterized in that, The M1 ports and / or the M2 ports are obtained by beamforming the N ports; or... The M1 ports and / or the M2 ports are obtained by sampling the N ports in the horizontal and / or vertical dimensions. or, The M1 ports are the first M1 ports out of the N ports, and / or the M2 ports are the first M2 ports out of the N ports; or, The M1 ports are the last M1 ports among the N ports, and / or the M2 ports are the last M2 ports among the N ports; or, The M1 ports and / or the M2 ports are obtained by uniformly sampling the N ports; or, The M1 ports and / or the M2 ports are sampled from the n port groups obtained by dividing the N ports, where n is a positive integer.

30. The method according to claim 23 or 24, characterized in that, The M1 ports and the M2 ports are two different sets of ports among the M3 ports. The first reference signal and the second reference signal are the same reference signal sent at the same time or time unit. M3 is a positive integer less than or equal to N.

31. The method according to claim 30, characterized in that, M3 is the sum of M1 and M2; The M1 ports are the first M1 ports out of the M3 ports, and the M2 ports are the last M2 ports out of the M3 ports; or, The M1 ports are the last M1 ports among the M3 ports, and the M2 ports are the first M2 ports among the M3 ports; or, The M1 ports are the ports with even indices among the M3 ports, and the M2 ports are the ports with odd indices among the M3 ports; or, The M1 ports are the ports with odd-numbered indices among the M3 ports, and the M2 ports are the ports with even-numbered indices among the M3 ports.

32. The method according to claim 30 or 31, characterized in that, The M3 ports are obtained by beamforming the N ports; or... The M3 ports are obtained by sampling the N ports in the horizontal and / or vertical dimensions; or... The M3 ports are the first M3 ports out of the N ports; or, The M3 ports are the last M3 ports among the N ports; or, The M3 ports are obtained by uniformly sampling the N ports; or, The M3 ports are sampled from the n port groups obtained by dividing the N ports, where n is a positive integer.

33. The method according to any one of claims 23 to 32, characterized in that, The method further includes: Send a third reference signal for M4 ports, the third reference signal for M4 ports being used in conjunction with the first reference signal for M1 ports. The first channel information is obtained by examining the signal; Wherein, the first reference signal is transmitted on the first frequency domain resource, the second reference signal is transmitted on the second frequency domain resource, the third reference signal is transmitted on the third frequency domain resource, and M4 is a positive integer less than or equal to N.

34. The method according to claim 33, characterized in that, The first frequency domain resource and the third frequency domain resource correspond to different bandwidths; or, The first frequency domain resource and the third frequency domain resource contain different numbers of subbands; or, The first frequency domain resource and the third frequency domain resource contain different numbers of subcarriers; or, The first frequency domain resource and the third frequency domain resource contain different numbers of Physical Resource Blocks (PRBs).

35. The method according to claim 33 or 34, characterized in that, The bandwidth corresponding to the first frequency domain resource is greater than the bandwidth corresponding to the third frequency domain resource; and / or, The bandwidth corresponding to the first frequency domain resource is greater than the bandwidth corresponding to the second frequency domain resource.

36. The method according to any one of claims 33 to 35, characterized in that, The subbands included in the second frequency domain resource are different from the subbands included in the third frequency domain resource; or, The subcarriers included in the second frequency domain resource are different from the subcarriers included in the third frequency domain resource; or, The PRBs included in the second frequency domain resource are different from those included in the third frequency domain resource.

37. The method according to any one of claims 33 to 36, characterized in that, The M4 ports are the N ports, or the M4 ports are a subset of the N ports whose indices differ from the M1 ports.

38. The method according to any one of claims 23 to 37, characterized in that, The second channel information is compared with the channel information of the corresponding M2 ports in the first channel information, and the comparison result is used to verify the first channel information.

39. The method according to claim 38, characterized in that, The comparison result is used to characterize the squared generalized cosine similarity (SGCS) between the second channel information and the channel information of the corresponding M2 ports in the first channel information; or, The comparison result is used to characterize the difference between the second channel information and the channel information of the corresponding M2 ports in the first channel information; or, The comparison result is used to characterize the correlation between the second channel information and the channel information of the corresponding M2 ports in the first channel information.

40. The method according to any one of claims 23 to 39, characterized in that, When M2 equals N, the transmission overhead of the second reference signal is negatively correlated with the interval duration for measuring the second channel information.

41. The method according to any one of claims 23 to 40, characterized in that, The verification is triggered by at least one of the following methods: Triggered by the second device either explicitly or implicitly; Triggered by the first device through a request.

42. The method according to any one of claims 23 to 41, characterized in that, If the verification result of the first channel information is inaccurate or invalid, the method further includes at least one of the following: Update the parameters of the AI / ML model that outputs the first channel information; Switch to the reference signal measurement results based on the N ports to obtain the channel information corresponding to the N ports; Switch the AI / ML model that outputs the first channel information.

43. The method according to any one of claims 23 to 42, characterized in that, The first channel information is output by the AI / ML model, and it is not outdated if the following conditions are met: The time interval between the moment when the AI / ML model outputs the first channel information and the moment when the first channel information is verified is less than or equal to the first interval duration. or, The time when the AI / ML model outputs the first channel information and the time when the first channel information is verified are within the same time window; or, The first channel information did not exceed the validity period.

44. The method according to any one of claims 23 to 43, characterized in that, The method used to obtain the first channel information and / or the second channel information is configured by the network device itself, or by the network device according to the capabilities of the first device.

45. A verification device, characterized in that, The device includes: The processing module is used to obtain first channel information based on the first reference signals of M1 ports. The first channel information is the channel information corresponding to N ports, where N is an integer greater than 1 and M1 is a positive integer less than N. The processing module is further configured to obtain second channel information based on the second reference signals of the M2 ports, wherein the second channel information is the channel information corresponding to the M2 ports, and M2 is a positive integer less than or equal to N; The processing module is further configured to verify the first channel information based on the second channel information.

46. ​​A verification device, characterized in that, The device includes: The transmitting module is used to transmit first reference signals for M1 ports. The first reference signals for M1 ports are used to determine first channel information. The first channel information is the channel information corresponding to N ports, where N is an integer greater than 1 and M1 is a positive integer less than N. The transmitting module is further configured to transmit second reference signals for M2 ports, the second reference signals for M2 ports being used to determine second channel information, the second channel information being the channel information corresponding to the M2 ports, where M2 is a positive integer less than or equal to N; The second channel information is used to verify the first channel information.

47. A communication device, characterized in that, The communication device includes a processor and a memory, the memory storing a computer program, the processor executing the computer program to implement the method as claimed in any one of claims 1 to 22, or to implement the method as claimed in any one of claims 23 to 44.

48. A computer-readable storage medium, characterized in that, The storage medium stores a computer program that is executed by a processor to implement the method as described in any one of claims 1 to 22, or to implement the method as described in any one of claims 23 to 44.

49. A chip, characterized in that, The chip includes programmable logic circuitry and / or program instructions, which, when the chip is running, are used to implement the method as described in any one of claims 1 to 22, or to implement the method as described in any one of claims 23 to 44.

50. A computer program product, characterized in that, The computer program product includes computer instructions stored in a computer-readable storage medium, which a processor reads from and executes to implement the method as claimed in any one of claims 1 to 22, or the method as claimed in any one of claims 23 to 44.

Images