Wireless communication method and communication device

Abstract

Provided are a wireless communication method and a communication device. The method comprises: on the basis of a first reference signal sent by a first device, a second device obtains first channel information corresponding to N antenna ports and / or a first frequency domain resource at a moment t1; on the basis of a second reference signal sent by the first device, the second device obtains second channel information corresponding to M antenna ports and / or a second frequency domain resource at a moment t2, wherein M<N, and the moment t2 follows the moment t1; and on the basis of the first channel information and the second channel information, the second device determines CSI corresponding to the N antenna ports and / or the first frequency domain resource at the moment t2 or a moment t3, wherein the moment t3 follows the moment t2.

Description

Wireless communication method and communication device TECHNICAL FIELD

[0001] The present application relates to the field of communication technology, and more particularly, to a wireless communication method and a communication device. BACKGROUND

[0002] At present, if complete channel state information (CSI) of uplink or downlink is to be obtained, the sending end needs to send reference signals corresponding to complete antenna ports (usually corresponding to the number of antennas of the sending end) and complete bandwidth, so that the receiving end obtains complete channel information by measurement, thereby obtaining the CSI of uplink or downlink and indicating it to the sending end. When the number of antennas of the sending end is large, the number of antenna ports is large and / or the CSI measurement bandwidth is large, frequent sending of reference signals of complete antenna ports and / or complete bandwidth requires a large amount of reference signal resources, which will affect the data transmission rate of uplink and downlink.

[0003] At present, with the aid of artificial intelligence (AI) / machine learning (ML) technology, the CSI on complete antenna ports or complete bandwidth can be recovered from channel information on partial antenna ports or partial bandwidth, thereby reducing the overhead of reference signals. However, due to various antenna configurations of network devices and various channels between network devices and terminal devices, if a single AI model is used to support all configurations / channels, the recovery performance will be poor under a specific configuration / channel. If different AI models are trained for different antenna configuration / channel scenarios, the terminal device and / or the network device need to frequently perform performance monitoring and switching of the models, thereby significantly increasing the complexity of the terminal device and / or the network device and the signaling overhead of uplink and downlink.

[0004] SUMMARY

[0005] The present application provides a wireless communication method and a communication device. Each aspect related to the present application is introduced below.

[0006] In a first aspect, a wireless communication method is provided, including: obtaining, by a second device, first channel information corresponding to N antenna ports and / or first frequency domain resources at time t1 based on a first reference signal sent by a first device, wherein the first reference signal corresponds to the N antenna ports and / or occupies the first frequency domain resources; obtaining, by the second device, second channel information corresponding to M antenna ports and / or second frequency domain resources at time t2 based on a second reference signal sent by the first device, wherein the second reference signal corresponds to the M antenna ports and / or occupies the second frequency domain resources, M < N, the second frequency domain resources are part of the first frequency domain resources, and time t2 is after time t1; and determining, by the second device, channel state information (CSI) corresponding to the N antenna ports and / or the first frequency domain resources at time t2 or time t3 according to the first channel information and the second channel information, wherein time t3 is after time t2.

[0007] In a second aspect, a wireless communication method is provided, including: sending, by a first device, a first reference signal at time t1, wherein the first reference signal corresponds to N antenna ports and / or occupies first frequency domain resources; sending, by the first device, a second reference signal at time t2, wherein the second reference signal corresponds to M antenna ports and / or occupies second frequency domain resources, M < N, and time t2 is after time t1; and receiving, by the first device, CSI sent by a second device, or receiving, by the first device, precoded data sent by the second device, wherein the precoded data is data precoded based on the CSI, the CSI corresponds to the N antenna ports and / or the first frequency domain resources at time t2 or time t3, the CSI is obtained based on the first reference signal and the second reference signal, and time t3 is after time t2.

[0008] In a third aspect, a communication device is provided, which is a second device, including: an obtaining unit, configured to obtain first channel information corresponding to N antenna ports and / or first frequency domain resources at time t1 based on a first reference signal sent by a first device, wherein the first reference signal corresponds to the N antenna ports and / or occupies the first frequency domain resources; the obtaining unit is further configured to obtain second channel information corresponding to M antenna ports and / or second frequency domain resources at time t2 based on a second reference signal sent by the first device, wherein the second reference signal corresponds to the M antenna ports and / or occupies the second frequency domain resources, M < N, the second frequency domain resources are part of the first frequency domain resources, and time t2 is after time t1; and a determining unit, configured to determine channel state information (CSI) corresponding to the N antenna ports and / or the first frequency domain resources at time t2 or time t3 according to the first channel information and the second channel information, wherein time t3 is after time t2.

[0009] In a fourth aspect, a communication device is provided, which is a first device, comprising: a sending unit configured to send a first reference signal at time t1, the first reference signal corresponding to N antenna ports and / or occupying a first frequency domain resource; the sending unit is further configured to send a second reference signal at time t2, the second reference signal corresponding to M antenna ports and / or occupying a second frequency domain resource, where M < N, and time t2 is after time t1; a receiving unit configured to receive CSI sent by a second device, or the receiving unit is configured to receive precoded data sent by the second device, the precoded data being data precoded based on the CSI, the CSI corresponding to the N antenna ports and / or the first frequency domain resource at time t2 or time t3, the CSI being obtained based on measurement of the first reference signal and the second reference signal, where time t3 is after time t2.

[0010] In a fifth aspect, a communication device is provided, which is a second device, comprising a processor, a memory, and a communication interface, the memory being configured to store one or more computer programs, and the processor being configured to invoke the computer programs in the memory, so that the terminal device performs part or all of the steps in the method of the first aspect.

[0011] In a sixth aspect, a communication device is provided, which is a first device, comprising a processor, a memory, and a transceiver, the memory being configured to store one or more computer programs, and the processor being configured to invoke the computer programs in the memory, so that the network device performs part or all of the steps in the method of the second aspect.

[0012] In a seventh aspect, an embodiment of the present application provides a communication system, which comprises the first device and / or the second device described above. In another possible design, the system can further comprise other devices interacting with the first device or the second device in the solutions provided by the embodiments of the present application.

[0013] In an eighth aspect, an embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program causes a communication device (for example, the first device or the second device) to perform part or all of the steps in the methods of the aspects described above.

[0014] In a ninth aspect, an embodiment of the present application provides a computer program product, which includes a non-transitory computer readable storage medium storing a computer program, and the computer program is operable to cause a communication device (for example, the first device or the second device) to perform part or all of the steps in the methods of the aspects described above. In some implementations, the computer program product can be a software installation package.

[0015] In a tenth aspect, an embodiment of the present application provides a chip, which comprises a memory and a processor. The processor can call and run a computer program from the memory to implement part or all of the steps described in the method of each of the above aspects.

[0016] Based on the above technical solution, when the second device obtains the CSI corresponding to the complete antenna ports (i.e., N antenna ports) and / or the complete bandwidth (i.e., the first frequency domain resource) at the time t2 or the time t3, the first device only needs to send the reference signal on the partial antenna ports (i.e., M antenna ports) or the partial bandwidth (i.e., the second frequency domain resource) at the time t2. The second device can obtain the CSI corresponding to the complete antenna ports and / or the complete bandwidth at the time t2 or the time t3 based on the first channel information at the time before the time t2 (i.e., the time t1) and the second channel information at the time t2. In this way, when the CSI corresponding to the complete antenna ports and / or the complete bandwidth at the time t2 or the time t3 is obtained, the first device does not need to send the reference signal corresponding to the complete antenna ports and / or the complete bandwidth at the time t2, thereby reducing the transmission overhead of the reference signal. In addition, since the statistical information of the complete channel is obtained by means of the first channel information with a low frequency, the model generalization and matching problems when the AI model is used to restore the complete channel information can be solved, thereby greatly improving the reliability of restoring the CSI in various different scenarios and configurations. BRIEF DESCRIPTION OF DRAWINGS

[0017] FIG. 1 is a wireless communication system 100 to which embodiments of the present application are applied.

[0018] FIG. 2 is a schematic diagram of different periodic CSI reporting methods provided by an embodiment of the present application.

[0019] FIG. 3 is a structure diagram of a neuron provided by an embodiment of the present application.

[0020] FIG. 4 is a structure diagram of a fully connected neural network provided by an embodiment of the present application.

[0021] FIG. 5 is a wireless communication method provided by an embodiment of the present application.

[0022] FIG. 6 is a schematic diagram of obtaining M antenna ports based on horizontal dimension sampling provided by an embodiment of the present application.

[0023] FIG. 7 is a schematic diagram of obtaining M antenna ports based on vertical dimension sampling provided by an embodiment of the present application.

[0024] FIG. 8 is a schematic diagram of obtaining M antenna ports based on horizontal dimension and vertical dimension sampling respectively provided by an embodiment of the present application.

[0025] FIG. 9 is a schematic diagram of the first M antenna ports provided by an embodiment of the present application.

[0026] FIG. 10 is a schematic diagram of the last M antenna ports according to an embodiment of the present application.

[0027] FIG. 11 is a schematic diagram of the first M / 2 groups of dual-polarized antennas according to an embodiment of the present application.

[0028] FIG. 12 is a schematic block diagram of a first device according to an embodiment of the present application.

[0029] FIG. 13 is a schematic block diagram of a second device according to an embodiment of the present application.

[0030] FIG. 14 is a schematic structural diagram of an apparatus according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] The technical solutions in the present application will be described below with reference to the accompanying drawings.

[0032] FIG. 1 is a wireless communication system 100 to which embodiments of the present application are applied. The wireless communication system 100 can include a network device 110 and a terminal device 120. The network device 110 can be a device that communicates with the terminal device 120. The network device 110 can provide communication coverage for a specific geographic area and can communicate with the terminal device 120 located in the coverage area.

[0033] FIG. 1 exemplarily shows one network device and two terminals. Optionally, the wireless communication system 100 can include multiple network devices and each network device can include other numbers of terminal devices within its coverage, which is not limited in the embodiments of the present application.

[0034] Optionally, the wireless communication system 100 can further include a network controller, a mobile management entity, and other network entities, which are not limited in the embodiments of the present application.

[0035] It should be understood that the technical solutions in the embodiments of the present application can be applied to various communication systems, for example, a 5th generation (5G) system or new radio (NR), a long term evolution (LTE) system, an LTE frequency division duplex (FDD) system, an LTE time division duplex (TDD), and the like. The technical solutions provided by the present application can also be applied to future communication systems, such as a 6th generation mobile communication system, a satellite communication system, and the like.

[0036] The terminal device in the embodiments of the present application can also be referred to as a user equipment (UE), an access terminal, a user unit, a user station, a mobile station, a mobile station (MS), a mobile terminal (MT), a remote station, a remote terminal, a mobile device, a user terminal, a terminal, a wireless communication device, a user agent or a user apparatus. The terminal device in the embodiments of the present application can refer to a device that provides voice and / or data connectivity for a user, and can be used to connect people, things and machines, such as handheld devices with wireless connection functions, vehicle-mounted devices, etc. The terminal device in the embodiments of the present application can be a mobile phone, a tablet computer (Pad), a notebook computer, a palm computer, a mobile internet device (MID), a wearable device, a virtual reality (VR) device, an augmented reality (AR) device, a wireless terminal in industrial control, a wireless terminal in self driving, a wireless terminal in remote medical surgery, a wireless terminal in smart grid, a wireless terminal in transportation safety, a wireless terminal in smart city, a wireless terminal in smart home, etc. Optionally, the UE can be used to act as a base station. For example, the UE can act as a scheduling entity, which provides sidelink signals between UEs in V2X or D2D, etc. For example, a cellular phone and a car communicate with each other using sidelink signals. The cellular phone and the smart home device communicate with each other without relaying the communication signals through the base station.

[0037] The network device in the embodiments of the present application can be a device for communicating with a terminal device, which can also be referred to as an access network device or a radio access network device, such as a network device, which can be a base station. The network device in the embodiments of the present application can refer to a radio access network (RAN) node (or device) that accesses a terminal device to a wireless network. The base station can broadly cover various names in the following or be replaced by the following names, such as: Node B (NodeB), evolved Node B (eNB), next generation Node B (gNB), relay station, transmitting and receiving point (TRP), transmitting point (TP), master station MeNB, auxiliary station SeNB, multi-standard radio (MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), positioning node, etc. The base station can be a macro base station, a micro base station, a relay node, a donor node or the like, or a combination thereof. The base station can also refer to a communication module, modem or chip used in the foregoing devices or apparatuses. The base station can also be a mobile switching center and a device that undertakes the function of a base station in device-to-device (D2D), vehicle-to-everything (V2X), machine-to-machine (M2M) communication, network side device in 6G network, device that undertakes the function of a base station in future communication systems, etc. The base station can support networks of the same or different access technologies. The embodiments of the present application do not limit the specific technology and specific device form adopted by the network device.

[0038] The base station can be fixed or mobile. For example, a helicopter or a drone can be configured to act as a mobile base station, and one or more cells can move according to the location of the mobile base station. In other examples, a helicopter or a drone can be configured to act as a device that communicates with another base station.

[0039] In some deployments, the network device in the embodiments of the present application can refer to a CU or a DU, or the network device includes a CU and a DU. The gNB can also include an AAU.

[0040] The network device and the terminal device can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; can also be deployed on water surface; can also be deployed on aircraft, balloons and satellites in the air. The scenarios in which the network device and the terminal device are located are not limited in the embodiments of the present application.

[0041] It should be understood that all or part of the functions of the communication device in the present application can also be implemented by software functions running on hardware, or by virtualized functions instantiated on a platform (such as a cloud platform).

[0042] CSI feedback

[0043] In order for the first device to reasonably schedule the second device, the second device needs to feed back downlink CSI to let the first device determine the scheduling information of the second device, such as the number of transmission layers, the precoding matrix, the transmission beam, the modulation and coding mode, etc. Further, the CSI reporting of the second device needs to be based on the CSI reporting configuration indicated by the first device and the channel state information-reference signal (CSI-RS) sent by the first device, and the uplink resource used by the second device to report CSI and the CSI-RS signal used for CSI measurement are both indicated by the CSI reporting configuration. Among them, each CSI reporting configuration corresponds to one CSI reporting, and each CSI reporting can contain different information, such as CSI-RS Resource Indicator (CRI), rank indicator (RI), precoding matrix indicator (PMI), channel quality indicator (CQI), etc., which are obtained based on the CSI-RS signal configured and sent by the first device. Specifically, which contents / information are contained in the CSI is determined by the report quantity information in the CSI reporting configuration. The report quantity information can indicate one of the following report quantities (extracted from 3GPP TS 38.331):

[0044] Among them, CRI is used to determine the CSI-RS resource currently used for channel measurement and the interference measurement resource (interference measurement resource, IMR) currently used for interference measurement from multiple CSI-RS resources; RI is used to report the recommended number of transmission layers; PMI is used to determine the recommended precoding matrix from the predefined codebook; and CQI is used to report the current channel quality.

[0045] Reference signal received power (RSRP) is used to report the RSRP of the synchronization signal block (synchronization signal block, SSB) or CSI-RS corresponding to the index fed back, so as to determine the beam used for downlink transmission by the first device.

[0046] Layer indicator (LI) is used to report the index of the transmission layer associated with the phase-tracking reference signal (phase-tracking reference signal, PTRS).

[0047] Among them, RI / PMI / CQI can be determined based on the signal to interference plus noise ratio (SINR) estimated by the terminal. The channel part in SINR is determined based on the non-zero power CSI-RS configured by the network for channel measurement, and the interference part is determined based on the channel state information (channel state information-interference measurement, CSI-IM) or non-zero power CSI-RS configured by the network for interference measurement. Among them, the CSI-RS resource for channel measurement can contain multiple antenna ports, which is used to measure the complete channel of the downlink to calculate the CSI.

[0048] There are three reporting modes for CSI reporting of a terminal: periodic CSI reporting, quasi-persistent CSI reporting and aperiodic CSI reporting, as shown in FIG. 2. Among them, the periodic CSI is transmitted on a physical uplink control channel (PUCCH), and the CSI reporting configuration thereof is configured by radio resource control (RRC), and the terminal periodically reports CSI after receiving the corresponding RRC configuration. The quasi-persistent CSI can be transmitted on the PUCCH or a physical uplink shared channel (PUSCH), the CSI reporting configuration corresponding to the CSI transmitted on the PUCCH is pre-configured by RRC signaling and activated or deactivated by medium access control (MAC) layer signaling, and the CSI reporting configuration corresponding to the CSI transmitted on the PUSCH is dynamically indicated (activated or deactivated) by downlink control information (DCI) signaling. After the terminal receives the activation signaling or indication signaling configured by the network, the CSI is periodically transmitted on the PUCCH or the PUSCH until the deactivation signaling is received to stop reporting. The aperiodic CSI reporting also has a pre-configured CSI reporting configuration through RRC signaling, part of the configuration can be activated by MAC layer signaling, and the CSI reporting configuration for CSI reporting is indicated by the CSI trigger signaling in the DCI. After the terminal receives the CSI trigger signaling, the corresponding CSI is reported on the scheduled PUSCH according to the indicated CSI reporting configuration.

[0049] Introduction of neural network

[0050] A neural network is an operation model composed of multiple neuron nodes connected to each other, wherein the connection between the nodes represents the weighted value from the input signal to the output signal, referred to as weight; each node performs weighted summation on different input signals and outputs through a specific activation function. The neuron structure is shown in FIG. 3.

[0051] A simple neural network is shown in FIG. 4, which includes an input layer, a hidden layer and an output layer, and can produce different outputs through different connection modes of multiple neurons, weights and activation functions, thereby fitting the mapping relationship from input to output. Each upper-level node is connected to all lower-level nodes. This full connection model can also be called a deep neural network (DNN).

[0052] A neural network model can be trained and obtained through the processes of dataset construction, training, verification and testing. Training can be divided into offline training and online training. The network can obtain a static training result through offline training of the dataset, which can be referred to as offline training. During the use of the neural network model by the network or the terminal, the network can continue to collect more data for real-time online training to optimize the parameters of the neural network model and achieve better inference and prediction results as the terminal further measures and / or reports. After obtaining the neural network model, the corresponding model output can be inferred by inputting the current obtained information into the model.

[0053] To implement different functions, different artificial intelligence (AI) models can be introduced, and corresponding inputs and outputs can be defined. For example, when AI / (machine learning, ML) is used for CSI feedback, the channel information (such as a feature vector, beam information, time delay information, etc.) obtained based on reference signal measurement can be used as the input of the model, so as to infer the corresponding CSI quantization bits. At the network side, there is a corresponding neural network model, which can infer the corresponding channel information by taking the CSI quantization bits as the input. For example, when AI / ML is used for data detection, that is, an AI receiver is used to detect a downlink signal, the terminal device can take the received signal (and some additional information such as a pilot sequence) as the input of the AI model for inference, so as to output the detected soft bit symbol (i.e., a constellation point). In addition, the AI model can also be used for positioning, channel coding, channel decoding, beam management, channel estimation and other processes.

[0054] Currently, if complete CSI of uplink or downlink is to be obtained, the sending end needs to send the corresponding reference signal on complete antenna ports (usually corresponding to the number of antennas of the receiving end) and / or complete bandwidth to the receiving end. After receiving the reference signal, the receiving end obtains complete channel information through measurement, thereby obtaining the CSI of uplink or downlink and indicating the sending end. When the number of antennas at the sending end is large, the number of antenna ports is large (for example, 128 / 256 ports for downlink and 16 ports for uplink), and / or the CSI measurement bandwidth is large, frequent sending of reference signals on complete antenna ports and / or complete bandwidth requires a large amount of reference signal resources, which not only increases the transmission power consumption of the sending end, but also affects the data transmission rate of uplink and downlink.

[0055] Currently, with the aid of AI / ML technology, the CSI on the complete antenna port or complete bandwidth can be recovered from the channel information of part of the antenna ports or part of the bandwidth, thereby reducing the overhead of the reference signal. However, due to the various antenna configurations of network devices and the channels between network devices and terminal devices, if a single AI model is used to support all configurations / channels, the recovery performance in a specific configuration / channel will be poor. If different AI models are trained for different antenna configuration / channel scenarios, the terminal device and / or the network device need to frequently monitor the performance of the models and switch, thereby significantly increasing the complexity of the terminal device and / or the network device and the signaling overhead of the uplink and downlink.

[0056] Based on this, the embodiments of the present application provide a wireless communication method and a communication device, which can obtain the CSI corresponding to the complete antenna port and / or complete frequency domain resource based on the reference signal corresponding to part of the antenna ports and / or part of the frequency domain resources, thereby reducing the transmission power consumption of the sending end and being beneficial to improving the data transmission rate of the uplink and downlink. For example, when the receiving end (such as the second device) obtains the CSI corresponding to the complete antenna port (i.e., N antenna ports) and / or complete bandwidth (i.e., the first frequency domain resource) at t2 or t3, the sending end (i.e., the first device) only needs to send the reference signal on part of the antenna ports (i.e., M antenna ports) or part of the bandwidth (i.e., the second frequency domain resource) at t2, and the second device can obtain the CSI corresponding to the complete antenna port and / or complete bandwidth at t2 or t3 based on the first channel information at a time point before t2 (i.e., t1) and the second channel information at t2. In this way, when the CSI corresponding to the complete antenna port and / or complete bandwidth at t2 or t3 is obtained, the first device does not need to send the reference signal corresponding to the complete antenna port and / or complete bandwidth at t2, thereby reducing the transmission overhead of the reference signal and reducing the power consumption of the first device. In addition, since the first channel information with low frequency is used to obtain the statistical information of the complete channel, the model generalization and matching problems when the AI model is used to recover the complete channel information are solved, and the reliability of recovering the CSI in various different scenarios and configurations is greatly improved.

[0057] The first device can be a first device for sending a reference signal, and the second device can be a second device for receiving a reference signal. In some embodiments, the first device can be a network device, and the second device can be a terminal device. In other embodiments, the first device can also be a terminal device, and the second device can also be a network device, which is suitable for different scenarios. For example, in the uplink data transmission scenario, the first device can be a terminal device, and the second device can be a network device. For another example, in the downlink data transmission scenario, the first device can be a network device, and the second device can be a terminal device.

[0058] The wireless communication method provided by the embodiment of the present application is introduced below in combination with FIG. 5. The method shown in FIG. 5 includes step S510.

[0059] In step S510, the second device determines the first channel information based on the first reference signal sent by the first device. The first channel information can correspond to N antenna ports and / or the first frequency domain resource at t1, where N is an integer greater than or equal to 1.

[0060] In some embodiments, the second device can obtain the first channel information based on the first reference signal sent by the first device. For example, the first device can send the first reference signal to the second device at t1, and the second device obtains the first channel information by measuring the first reference signal.

[0061] In some implementations, the first reference signal corresponds to N antenna ports. The first reference signal corresponding to N antenna ports can mean that the first reference signal is sent through N antenna ports, or that the first reference signal includes N antenna ports. N corresponds to the number of antennas of the second device, or in other words, the second device includes N antennas. The N antenna ports can also be referred to as complete antenna ports.

[0062] The embodiment of the present application does not make specific limitations on the value of the antenna port N. In some embodiments, the value of N is related to the application scenario. For example, in downlink data transmission, the value of N can be 32, can be 64, can be 128, can be 256, and can also be other values. For another example, in uplink data transmission, the value of N can be 8, can be 16, can be 32, and can also be other values.

[0063] In some implementations, the first reference signal occupies the first frequency domain resource. The first reference signal occupying the first frequency domain resource can mean that the first reference signal is sent through the first frequency domain resource. The first frequency domain resource can be referred to as complete bandwidth, or can also be referred to as complete frequency domain resource. The embodiment of the present application does not make specific limitations on the type of the first frequency domain resource. For example, the first frequency domain resource can include one or more bandwidths, or the first frequency domain resource can include one or more subbands, or the first frequency domain resource can include one or more physical resource blocks (PRBs), or the first frequency domain resource can include one or more subcarriers.

[0064] The type of the first reference signal is not limited in the embodiments of the present application. For example, the first reference signal can be a CSI-RS signal, corresponding to one CSI-RS resource. For another example, the first reference signal can also be a sounding reference signal (SRS), corresponding to one or more SRS resources. The SRS resource can be an SRS resource for antenna switching. For example, the first reference signal is transmitted through multiple SRS resources for antenna switching, the number of ports of different SRS resources can be the same and correspond to different antennas, and the ports of these SRS resources can be combined to obtain N antenna ports. For example, one N=8 antenna port reference signal is obtained through 4 2 antenna port SRS resources. These SRS resources can be transmitted in one time slot or adjacent time slots, and the t1 moment can correspond to one SRS sending window.

[0065] In some embodiments, the t1 moment can correspond to a time domain resource. The time domain resource can be configured for the first device or predefined by a protocol. For example, the t1 moment can correspond to a time domain resource configured for the first device to send the first reference signal, that is, the t1 moment is a time domain resource configured by the first device, used to send the first reference signal to the second device.

[0066] The content contained in the first channel information is not limited in the embodiments of the present application. As an example, the first channel information can include one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, value of N, line of sight (LOS) / non line of sight (NLOS) indication, channel multipath parameter.

[0067] In some embodiments, the first channel information can include a channel matrix. For example, the first channel information can include a K*N dimensional channel matrix, where K is the number of receive antennas of the second device (such as a UE).

[0068] In another embodiment, the first channel information can include a channel covariance matrix. For example, the first channel information can include an N*N dimensional channel covariance matrix.

[0069] In another embodiment, the first channel information can include a channel eigenvector. For example, the first channel information can include one or more N*1 dimensional channel eigenvectors.

[0070] In another embodiment, the first channel information can include indication information. For example, the first channel information can include an indication of LOS / NLOS, used to indicate whether the current measurement obtained channel is an LOS channel.

[0071] In some embodiments, the first channel information can comprise parameter information. For example, the first channel information can comprise the multi-path parameters of the channel, such as the time delay and power spectrum of the multi-paths, etc. In some implementations, the first channel information can comprise one or more of the following information: the time delay of each multi-path, the power spectrum of each multi-path, the maximum time delay of the channel of each multi-path.

[0072] In some embodiments, the first channel information can comprise the value of N. For example, the first channel information can further comprise the value of N on the basis of the aforementioned information.

[0073] In some embodiments, the method shown in FIG. 5 can further comprise step S520. In step S520, the second device determines the second channel information based on the second reference signal sent by the first device. The second channel information can correspond to the M antenna ports and / or the second frequency domain resource at time t2. M is an integer greater than or equal to 1, and M < N.

[0074] In some embodiments, the second device can obtain the second channel information based on the second reference signal sent by the first device. For example, the first device can send the second reference signal to the second device at time t2, and the second device can obtain the second channel information by measuring the second reference signal.

[0075] In some embodiments, the second device can send the first capability information to the first device before receiving the second reference signal sent by the first device, to indicate that the second device has the capability of performing CSI measurement based on the M antenna ports and / or the second frequency domain resource. That is, the capability indicates whether the second device can obtain the CSI corresponding to the current N antenna ports based on the reference signal of the N antenna ports sent by the first device before and the reference signal of the M antenna ports sent by the first device currently, or the capability indicates whether the second device can obtain the CSI corresponding to the current first frequency domain resource based on the reference signal of the first frequency domain resource sent by the first device before and the reference signal of the second frequency domain resource sent by the first device currently. Or, the capability indicates whether the second device can obtain the CSI corresponding to the current N antenna ports and first frequency domain resource based on the reference signal of the N antenna ports and first frequency domain resource sent by the first device before and the reference signal of the M antenna ports and second frequency domain resource sent by the first device currently. If the second device indicates that it has the capability, the reference signal resource overhead can be reduced by the method of the present application, otherwise, the first device needs to continuously send the reference signal of the N antenna ports and / or the first frequency domain resource.

[0076] In some embodiments, the second reference signal corresponds to M antenna ports. The second reference signal corresponding to M antenna ports can mean that the second reference signal is transmitted through the M antenna ports, or that the second reference signal includes the M antenna ports. M corresponds to the number of partial antennas of the second device. The M antenna ports can also be referred to as partial antenna ports.

[0077] In some embodiments, the second reference signal occupies a second frequency domain resource. The second reference signal occupying the second frequency domain resource can mean that the second reference signal is transmitted through the second frequency domain resource. The second frequency domain resource can also be referred to as a partial bandwidth, or also referred to as a partial frequency domain resource. The embodiments of the present application do not make specific limitations on the type of the first frequency domain resource. For example, the second frequency domain resource can include one or more bandwidths, or the second frequency domain resource can include one or more sub-bands, or the second frequency domain resource can include one or more PRBs, or the second frequency domain resource can include one or more subcarriers. The second frequency domain resource is a partial resource in the first frequency domain resource. For example, the second frequency domain resource is different from the first frequency domain resource and is a subset of the first frequency domain resource.

[0078] The embodiments of the present application do not make specific limitations on the type of the second reference signal. For example, the second reference signal can be a CSI-RS signal, corresponding to one CSI-RS resource. For another example, the second reference signal can be an SRS signal, corresponding to one or more SRS resources. Of course, the second reference signal can also be other types of signals, which will not be described here. It should be noted here that the CSI-RS resources and / or SRS resources corresponding to the first reference signal and the second reference signal can be different. For example, the CSI-RS resources corresponding to the first reference signal and the second reference signal can be different CSI-RS resources in the same CSI-RS resource set, or the SRS resources corresponding to the first reference signal and the second reference signal can be different SRS resources in the same SRS resource set.

[0079] In some embodiments, the t2 time point can correspond to a time domain resource. The time domain resource can be configured for the first device, or be pre-defined by the protocol. For example, the t2 time point can correspond to a time domain resource configured by the first device for transmitting the second reference signal, that is, the t2 time point is a time domain resource configured by the first device for transmitting the second reference signal. Wherein, the t2 time point is after the t1 time point, indicating that the time domain resource for transmitting the second reference signal is after the time domain resource for transmitting the first reference signal, that is, the second device receives the first reference signal first, and then receives the second reference signal.

[0080] The embodiments of the present application do not make specific limitations on the relationship between the first frequency domain resource and the second frequency domain resource.

[0081] For example, the first frequency domain resource and the second frequency domain resource can correspond to different bandwidths, that is, the bandwidth corresponding to the first frequency domain resource is different from the bandwidth corresponding to the second frequency domain resource. The data transmission capability of the first frequency domain resource is different from that of the second frequency domain resource. For example, the bandwidth contained by the second frequency domain resource is smaller than the bandwidth contained by the first frequency domain resource.

[0082] For another example, the first frequency domain resource and the second frequency domain resource contain different numbers of subbands. The number of subbands contained by the second frequency domain resource is smaller than the number of subbands contained by the first frequency domain resource. For example, the first frequency domain resource contains all subbands that need to measure CSI, and the second frequency domain resource contains part of the subbands that need to measure CSI. For example, the first frequency domain resource contains all subbands that need to measure CSI, and the second frequency domain resource contains odd-numbered subbands or even-numbered subbands that need to measure CSI, or the second frequency domain resource contains part of the subbands that are every interval of one or more subbands in the first frequency domain resource.

[0083] For another example, the first frequency domain resource and the second frequency domain resource can contain different numbers of subcarriers. For example, the second frequency domain resource contains fewer subcarriers than the first frequency domain resource. For example, the first frequency domain resource contains all subcarriers that need to measure CSI, and the second frequency domain resource contains part of the subcarriers that need to measure CSI. For example, the first frequency domain resource contains all subcarriers that need to measure CSI, and the second frequency domain resource contains odd-numbered subcarriers or even-numbered subcarriers that need to measure CSI, or the second frequency domain resource contains part of the subcarriers that are every interval of one or more subcarriers in the first frequency domain resource.

[0084] For another example, the first frequency domain resource and the second frequency domain resource can contain different numbers of physical resource blocks (PRBs). For example, the second frequency domain resource contains fewer PRBs than the first frequency domain resource. The first frequency domain resource contains all PRBs that need to measure CSI, and the second frequency domain resource contains part of the PRBs that need to measure CSI. For example, the first frequency domain resource contains all PRBs that need to measure CSI, and the second frequency domain resource contains odd-numbered PRBs or even-numbered PRBs that need to measure CSI, or the second frequency domain resource contains part of the PRBs that are every interval of one or more PRBs in the first frequency domain resource.

[0085] Embodiments of the present application do not specifically limit the content contained in the second channel information. As an example, the second channel information includes one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, and value of M.

[0086] In some embodiments, the second channel information can comprise a channel matrix. For example, the second channel information can comprise a channel matrix of K*M dimensions, where K is the number of receive antennas of the second device (e.g., a UE). For another example, the second channel information can comprise a channel matrix of K*M dimensions on the second frequency domain resource, where K is the number of receive antennas of the second device (e.g., a UE).

[0087] In some other embodiments, the second channel information can comprise a channel covariance matrix. For example, the second channel information can comprise a channel covariance matrix of M*M dimensions. For another example, the second channel information can comprise a channel covariance matrix of M*M dimensions on the second frequency domain resource.

[0088] In some other embodiments, the second channel information can comprise a channel eigenvector. For example, the second channel information can comprise one or more channel eigenvectors of M*1 dimensions. For another example, the second channel information can comprise one or more channel eigenvectors of M*1 dimensions on the second frequency domain resource.

[0089] In some other embodiments, the second channel information can comprise a value of M. For example, the second channel information can further comprise a value of M on the basis of the aforementioned information.

[0090] The method shown in FIG. 5 can further comprise step S530, in which the second device determines CSI. In some embodiments, the second device determines, according to the first channel information and the second channel information, CSI corresponding to the N antenna ports and / or the first frequency domain resource at time t2 or time t3, where time t3 is after time t2.

[0091] Embodiments of the present application do not make specific limitations on the information included in the CSI. In some embodiments, the CSI can comprise one or more of the following information: a precoding matrix, a channel matrix, and an eigenvector. Taking downlink transmission as an example, the CSI can comprise one or more of the following information: a precoding matrix for downlink, a downlink channel matrix, and a downlink eigenvector. Taking uplink transmission as an example, the CSI can comprise one or more of the following information: a precoding matrix for uplink, an uplink channel matrix, and an uplink eigenvector. In some embodiments, the CSI at least comprises PMI information. For example, the CSI can comprise PMI information, which can be used by the first device to obtain a precoding matrix for uplink and / or downlink transmission. In some embodiments, the CSI can comprise other information such as RI, CQI, MCS, etc.

[0092] In some embodiments, the second device can determine, according to the first channel information and the second channel information, CSI corresponding to the N antenna ports at time t2 or time t3, where time t3 is after time t2.

[0093] In some embodiments, the second device can determine the CSI corresponding to the first frequency domain resource at the t2 moment or the t3 moment according to the first channel information and the second channel information, where the t3 moment is after the t2 moment.

[0094] In some embodiments, the second device can determine the CSI corresponding to the N antenna ports and the first frequency domain resource at the t2 moment or the t3 moment according to the first channel information and the second channel information, where the t3 moment is after the t2 moment.

[0095] The following describes the ways in which the second device determines the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t2 moment or the t3 moment according to the first channel information and the second channel information.

[0096] There are many ways for the second device to determine the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t2 moment according to the first channel information and the second channel information, which are not limited by the embodiments of the present application. As an example, the second device can input the first channel information and the second channel information into an AI model to output the CSI corresponding to the t2 moment. As another example, the second device can determine an AI model based on the first channel information, and input the second channel information into the AI model to output the CSI corresponding to the t2 moment. As another example, the second device can input the first channel information and the second channel information into an AI model to output third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the t2 moment, and determine the CSI corresponding to the t2 moment based on the third channel information. As another example, the second device can determine an AI model based on the first channel information, and input the second channel information into the AI model to output third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the t2 moment, and determine the CSI corresponding to the t2 moment based on the third channel information. The following describes the above ways of determining the CSI.

[0097] In some embodiments, the second device can input the first channel information and the second channel information into an AI model to output the CSI corresponding to the t2 moment, including: the second device inputs the first channel information and the second channel information into a first AI model to output the CSI corresponding to the N antenna ports at the t2 moment, and / or the second device inputs the first channel information and the second channel information into the first AI model to output the CSI corresponding to the first frequency domain resource at the t2 moment.

[0098] In some implementations, the second device outputs, through the first AI model, the CSI corresponding to the N antenna ports at the t2 moment as an output of the first AI model taking the first channel information and the second channel information as an input. For example, the input of the first AI model can be the first channel information and the second channel information, and the output can be the CSI corresponding to the N antenna ports at the t2 moment, where the CSI can be CSI bits.

[0099] In some implementations, the second device outputs, through the first AI model, the CSI corresponding to the first frequency domain resource at the t2 moment as an output of the first AI model taking the first channel information and the second channel information as an input. For example, the input of the first AI model can be the first channel information and the second channel information, and the output can be the CSI corresponding to the first frequency domain resource at the t2 moment, where the CSI can be CSI bits.

[0100] In some implementations, the second device outputs, through the first AI model, the CSI corresponding to the N antenna ports and the first frequency domain resource at the t2 moment as an output of the first AI model taking the first channel information and the second channel information as an input. For example, the input of the first AI model can be the first channel information and the second channel information, and the output can be the CSI corresponding to the N antenna ports and the first frequency domain resource at the t2 moment, where the CSI can be CSI bits.

[0101] In some embodiments, the second device can determine an AI model based on the first channel information, and take the second channel information as an input of the AI model to output the corresponding CSI at the t2 moment, including: the second device determines a second AI model according to the first channel information, and takes the second channel information as an input of the second AI model, and outputs the CSI corresponding to the N antenna ports at the t2 moment through the second AI model; and / or, the second device determines a second AI model according to the first channel information, and takes the second channel information as an input of the second AI model, and outputs the CSI corresponding to the first frequency domain resource at the t2 moment through the second AI model.

[0102] In some implementations, the second device determines a second AI model according to the first channel information, and takes the second channel information as an input of the second AI model, and outputs the CSI corresponding to the N antenna ports at the t2 moment through the second AI model. For example, the second device can first determine a second AI model according to the first channel information, and then take the second channel information as an input of the second AI model, and output the CSI corresponding to the N antenna ports at the t2 moment through the second AI model, where the CSI can be CSI bits.

[0103] In some implementations, the second device determines the second AI model according to the first channel information, and inputs the second channel information into the second AI model to output the CSI corresponding to the first frequency domain resource at the time t2 via the second AI model. For example, the second device can first determine the second AI model according to the first channel information, and then input the second channel information into the second AI model to output the CSI corresponding to the first frequency domain resource at the time t2 via the second AI model. The CSI can be CSI bits.

[0104] In some implementations, the second device determines the second AI model according to the first channel information, and inputs the second channel information into the second AI model to output the CSI corresponding to the first frequency domain resource at the time t2 via the second AI model. For example, the second device can first determine the second AI model according to the first channel information, and then input the second channel information into the second AI model to output the CSI corresponding to the first frequency domain resource at the time t2 via the second AI model. The CSI can be CSI bits.

[0105] There are various ways for the second device to determine the second AI model according to the first channel information, which are not limited in the embodiments of the present application. As an example, the second device can determine the second AI model according to the value of the first channel information. Different values of the first channel information can correspond to different AI models, and the second device can determine the second AI model from multiple candidate models according to the value of the first channel information. The value of the first channel can include one or more of the following: LOS / NLOS, value of N, number of multipaths, maximum multipath delay, etc. As another example, the second device can determine the second AI model according to the first channel information and other AI models (such as the ninth AI model). The second device can input the first channel information into the ninth AI model to determine the second AI model. The ninth AI model is used to determine the second AI model, and the function of the ninth AI model is to determine the second AI model from multiple candidate models. For example, the second device can input the measured time delay and power values of multiple paths into the model to determine the second AI model.

[0106] After the second device determines the second AI model, the second device can input the second channel information into the second AI model to obtain the output of the second AI model. The output of the second model is the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the time t2.

[0107] In some embodiments, the second device can input the first channel information and the second channel information as inputs of the AI model to output third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the time t2, and determine the CSI corresponding to the time t2 based on the third channel information, including: the second device inputs the first channel information and the second channel information as inputs of a third AI model to output the third channel information corresponding to the N antenna ports at the time t2 through the third AI model, and determines the CSI corresponding to the N antenna ports at the time t2 according to the third channel information, and / or the second device inputs the first channel information and the second channel information as inputs of the third AI model to output the third channel information corresponding to the first frequency domain resource at the time t2 through the third AI model, and determines the CSI corresponding to the first frequency domain resource at the time t2 according to the third channel information

[0108] In some implementations, the second device inputs the first channel information and the second channel information as inputs of the third AI model to output the third channel information corresponding to the N antenna ports at the time t2 through the third AI model, and determines the CSI corresponding to the N antenna ports at the time t2 according to the third channel information. For example, the second device inputs the first channel information and the second channel information into the third AI model to output the third channel information corresponding to the N antenna ports at the time t2, and determines the CSI corresponding to the N antenna ports at the time t2 according to the third channel information after the second device obtains the third channel information.

[0109] It should be noted that the third channel information is not the direct CSI bit, but the channel information similar to the second channel information. The difference between the third channel information and the second channel information is the number of corresponding antenna ports. The second channel information corresponds to part of the channel information of the M antenna ports, and the third channel information corresponds to the complete channel information of the N antenna ports.

[0110] There are many methods for the second device to calculate the CSI corresponding to the N antenna ports at the time t2 based on the third channel information, which are not limited in the present application. In some embodiments, the method for the second device to calculate the CSI corresponding to the N antenna ports at the time t2 based on the third channel information can reuse the method of related technologies, that is, the third channel information is equivalent to the channel information measured by the reference signal of the N antenna ports in related technologies. Since only the second reference signal of the M antenna ports is needed, the overhead of the reference signal can be reduced.

[0111] In some implementations, the second device inputs the first channel information and the second channel information as inputs of the third AI model, to output, by the third AI model, third channel information corresponding to the first frequency domain resource at the time t2, and determine the CSI corresponding to the first frequency domain resource at the time t2 according to the third channel information. For example, the second device inputs the first channel information and the second channel information into the third AI model, to output the third channel information corresponding to the first frequency domain resource at the time t2, and determine the CSI corresponding to the first frequency domain resource at the time t2 according to the third channel information after the second device obtains the third channel information.

[0112] It should be noted that the third channel information is not the direct CSI bits, but the channel information similar to the second channel information. The difference between the third channel information and the second channel information is that the corresponding frequency domain resources are different. For example, the second channel information corresponds to part of the channel information of the first frequency domain resource, and the third channel information is the complete channel information corresponding to the first frequency domain resource.

[0113] In some implementations, the second device inputs the first channel information and the second channel information as inputs of the third AI model, to output, by the third AI model, third channel information corresponding to the first frequency domain resource at the time t2, and determine the CSI corresponding to the first frequency domain resource at the time t2 according to the third channel information. For example, the second device inputs the first channel information and the second channel information into the third AI model, to output the third channel information corresponding to the first frequency domain resource at the time t2, and determine the CSI corresponding to the first frequency domain resource at the time t2 according to the third channel information after the second device obtains the third channel information.

[0114] It should be noted that the third channel information is not the direct CSI bits, but the channel information similar to the second channel information. The difference between the third channel information and the second channel information is that the corresponding N antenna ports and frequency domain resources are different. For example, the second channel information corresponds to M antenna ports and the second frequency domain resource, and the third channel information corresponds to N antenna ports and the first frequency domain resource.

[0115] There are many methods for the second device to calculate the CSI corresponding to the first frequency domain resource at the time t2 based on the third channel information, which are not limited in the present application. In some embodiments, the method for the second device to calculate the CSI corresponding to the first frequency domain resource at the time t2 based on the third channel information can reuse the method of related technologies, that is, the third channel information is equivalent to the channel information measured by the reference signal on the first frequency domain resource in related technologies. Since only the second reference signal on the second frequency domain resource is needed, the overhead of the reference signal can be reduced.

[0116] The third channel information includes one or more of the following: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of N.

[0117] In some embodiments, the third channel information can include a channel matrix. For example, the third channel information can include a channel matrix of K*N dimensions, where K is the number of receive antennas of the second device (e.g., a UE).

[0118] In some other embodiments, the third channel information can include a channel covariance matrix. For example, the third channel information can include a channel covariance matrix of N*N dimensions.

[0119] In some other embodiments, the third channel information can include a channel eigenvector. For example, the third channel information can include one or more channel eigenvectors of M*1 dimensions. For another example, the third channel information can be the same type of channel information as the second channel information, e.g., both the third channel information and the second channel information are channel eigenvectors.

[0120] In some other embodiments, the third channel information can include a value of N. For example, the third channel information can further include a value of N on the basis of the aforementioned information.

[0121] In some embodiments, the second device can determine an AI model based on the first channel information, and input the second channel information into the AI model to output third channel information corresponding to the N antenna ports and / or the first frequency domain resource at time t2, and determine the corresponding CSI at time t2 based on the third channel information, including: the second device determines a fourth AI model according to the first channel information, and inputs the second channel information into the fourth AI model to output third channel information corresponding to the N antenna ports at time t2 through the fourth AI model, and determines the CSI corresponding to the N antenna ports at time t2 according to the third channel information, and / or the second device determines a fourth AI model according to the first channel information, and inputs the second channel information into the fourth AI model to output third channel information corresponding to the first frequency domain resource at time t2 through the fourth AI model, and determines the CSI corresponding to the first frequency domain resource at time t2 according to the third channel information.

[0122] In some implementations, the second device determines a fourth AI model according to the first channel information, and inputs the second channel information as an input of the fourth AI model to output third channel information corresponding to the N antenna ports at the t2 moment through the fourth AI model. The second device can determine the CSI corresponding to the N antenna ports at the t2 moment according to the third channel information. For example, the second device can first determine the fourth AI model according to the first channel information, then input the second channel information as the input of the fourth AI model, obtain the output of the fourth AI model, that is, the third channel information corresponding to the N antenna ports at the t2 moment, and finally determine the CSI corresponding to the N antenna ports at the t2 moment according to the third channel information.

[0123] In some implementations, the second device determines a fourth AI model according to the first channel information, and inputs the second channel information as an input of the fourth AI model to output third channel information corresponding to the first frequency domain resource at the t2 moment. The second device can determine the CSI corresponding to the first frequency domain resource at the t2 moment according to the third channel information. For example, the second device can first determine the fourth AI model according to the first channel information, then input the second channel information as the input of the fourth AI model, obtain the output of the fourth AI model, that is, the third channel information corresponding to the first frequency domain resource at the t2 moment, and finally determine the CSI corresponding to the first frequency domain resource at the t2 moment according to the third channel information.

[0124] In some implementations, the second device determines a fourth AI model according to the first channel information, and inputs the second channel information as an input of the fourth AI model to output third channel information corresponding to the N antenna ports and the first frequency domain resource at the t2 moment. The second device can determine the CSI corresponding to the N antenna ports and the first frequency domain resource at the t2 moment according to the third channel information. For example, the second device can first determine the fourth AI model according to the first channel information, then input the second channel information as the input of the fourth AI model, obtain the output of the fourth AI model, that is, the third channel information corresponding to the N antenna ports and the first frequency domain resource at the t2 moment, and finally determine the CSI corresponding to the N antenna ports and the first frequency domain resource at the t2 moment according to the third channel information.

[0125] The method for determining the fourth AI model by the second device is similar to the method for determining the second AI model described above, which will not be described here.

[0126] The embodiments of the present application do not make specific limitations on the method of training the AI model by the second device. In some embodiments, the second device can train the AI model based on the collected training data set. The collected data includes data collected based on different antenna configurations, different antenna mapping methods and / or different channel environments. Through the above training process, the trained AI model can be applicable to different antenna configurations, antenna mapping methods and / or channel environments. For example, the AI model can identify the current antenna configuration, antenna mapping method and / or channel environment through the first channel information, so as to determine the AI model to be used. The second device can obtain the information of antenna configuration, antenna mapping method and / or channel environment based on the complete first channel information, so as to complete the recovery process of the second channel information to the complete channel information without the configuration and assistance of the first device. In this case, the first device and the second device can use the same model, which can avoid the interaction of the models between the first device and the second device, thereby saving a large amount of signaling and time related to model management.

[0127] In other embodiments, the second device can train the AI model based on the training data set collected in different antenna configurations, so that the trained AI model can be applicable to different antenna configurations. For example, the AI model can identify the current antenna configuration through the first channel information, so as to determine the AI model to be used according to the antenna configuration determined by the first channel information.

[0128] In other embodiments, the second device can train the AI model based on the training data set collected in different antenna mapping methods, so that the trained AI model can be applicable to different antenna mapping methods. For example, the AI model can identify the current antenna mapping method through the first channel information, so as to determine the AI model to be used according to the antenna mapping method determined by the first channel information.

[0129] In other embodiments, the second device can train the AI model based on the training data set collected in different channel environments, so that the trained AI model can be applicable to different channel environments. For example, the AI model can identify the current channel environment through the first channel information, so as to determine the AI model to be used according to the channel environment determined by the first channel information.

[0130] The following describes the way in which the second device determines the CSI corresponding to the N antenna ports and / or the first frequency domain resource at t3 according to the first channel information and the second channel information.

[0131] There are many ways for the second device to determine the CSI corresponding to the N antenna ports and / or the first frequency domain resource at time t3 (hereinafter referred to as the CSI corresponding to time t2) according to the first channel information and the second channel information, and embodiments of the present application do not make specific limitations thereon. As an example, the second device can take the first channel information and the second channel information as inputs of an AI model to output the CSI corresponding to time t3. As another example, the second device can determine an AI model based on the first channel information, and take the second channel information as an input of the AI model to output the CSI corresponding to time t3. As another example, the second device can take the first channel information and the second channel information as inputs of an AI model to output third channel information corresponding to the N antenna ports and / or the first frequency domain resource at time t3, and determine the CSI corresponding to time t3 based on the third channel information. As another example, the second device can determine an AI model based on the first channel information, and take the second channel information as an input of the AI model to output third channel information corresponding to the N antenna ports and / or the first frequency domain resource at time t3, and determine the CSI corresponding to time t3 based on the third channel information. The above-mentioned ways of determining the CSI corresponding to time t3 are introduced respectively as follows.

[0132] In some embodiments, the second device takes the first channel information and the second channel information as inputs of an AI model to output the CSI corresponding to time t3, including: the second device takes the first channel information and the second channel information corresponding to the M antenna ports at multiple different times as inputs of a fifth AI model to output the CSI corresponding to the N antenna ports at time t3 through the fifth AI model. Wherein, the second channel information corresponding to the M antenna ports at multiple different times includes the second channel information corresponding to the M antenna ports at time t2.

[0133] In some embodiments, the second device takes the first channel information and the second channel information as inputs of an AI model to output the CSI corresponding to time t3, can include: the second device takes the first channel information and the second channel information corresponding to the second frequency domain resource at multiple different times as inputs of a fifth AI model to output the CSI corresponding to the second frequency domain resource at time t3 through the fifth AI model. Wherein, the second channel information corresponding to the second frequency domain resource at multiple different times includes the second channel information corresponding to the second frequency domain resource at time t2.

[0134] In some embodiments, the second device inputs the first channel information and the second channel information as inputs of the AI model to output the CSI corresponding to the time t3, including: the second device inputs the first channel information and the second channel information corresponding to the M antenna ports and the second frequency domain resource at multiple different times as inputs of a fifth AI model, to output the CSI corresponding to the N antenna ports and the first frequency domain resource at the time t3 through the fifth AI model. Wherein, the second channel information corresponding to the M antenna ports and the second frequency domain resource at multiple different times includes the second channel information corresponding to the M antenna ports and the second frequency domain resource at the time t2.

[0135] In some embodiments, the second channel information corresponding to the M antenna ports at multiple different times can include the second channel information corresponding to the M antenna ports at any one or more time instants between the time t1 and the time t3. Wherein, the one or more time instants can include the time t2.

[0136] In some embodiments, the second channel information corresponding to the second frequency domain resource at multiple different times can include the second channel information corresponding to the second frequency domain resource at any one or more time instants between the time t1 and the time t3. Wherein, the one or more time instants can include the time t2.

[0137] In some embodiments, the second channel information corresponding to the M antenna ports and the second frequency domain resource at multiple different times can include the second channel information corresponding to the M antenna ports and the second frequency domain resource at any one or more time instants between the time t1 and the time t3. Wherein, the one or more time instants can include the time t2.

[0138] In some implementations, the second device inputs the first channel information and the second channel information corresponding to the M antenna ports at multiple different times as inputs of the fifth AI model to output the CSI corresponding to the N antenna ports at the time t3 through the fifth AI model. That is, the second device can determine the CSI corresponding to the N antenna ports at the time t3 according to the first channel information and the second channel information corresponding to the M antenna ports at multiple different times. For example, the input of the fifth AI model can be the first channel information and the second channel information corresponding to the M antenna ports at multiple different times, and the output can be the CSI corresponding to the N antenna ports at the time t3. Wherein, the CSI can be CSI bits.

[0139] In some implementations, the second device takes the first channel information and the second channel information corresponding to the second frequency domain resource at multiple different time instants as inputs of the fifth AI model, to output the CSI corresponding to the first frequency domain resource at t3 through the fifth AI model. That is, the second device can determine the CSI corresponding to the first frequency domain resource at t3 according to the first channel information and the second channel information corresponding to the second frequency domain resource at multiple different time instants. For example, the input of the fifth AI model can be the first channel information and the second channel information corresponding to the second frequency domain resource at multiple different time instants, and the output can be the CSI corresponding to the first frequency domain resource at t3. Wherein, the CSI can be CSI bits.

[0140] In some implementations, the second device takes the first channel information and the second channel information corresponding to the second frequency domain resource at multiple different time instants as inputs of the fifth AI model, to output the CSI corresponding to the first frequency domain resource at t3 through the fifth AI model. That is, the second device can determine the CSI corresponding to the first frequency domain resource at t3 according to the first channel information and the second channel information corresponding to the second frequency domain resource at multiple different time instants. For example, the input of the fifth AI model can be the first channel information and the second channel information corresponding to the second frequency domain resource at multiple different time instants, and the output can be the CSI corresponding to the first frequency domain resource at t3. Wherein, the CSI can be CSI bits.

[0141] In some embodiments, the second device determines an AI model based on the first channel information, and takes the second channel information as an input of the AI model to output the corresponding CSI at t2, which can include: the second device determines a sixth AI model according to the first channel information, and takes the second channel information corresponding to the M antenna ports at multiple different time instants as an input of the sixth AI model, to output the CSI corresponding to the N antenna ports at t3 through the sixth AI model. The CSI can be CSI bits.

[0142] In some embodiments, the second device determines an AI model based on the first channel information, and takes the second channel information as an input of the AI model to output the corresponding CSI at t2, which can include: the second device determines a sixth AI model according to the first channel information, and takes the second channel information corresponding to the second frequency domain resource at multiple different time instants as an input of the sixth AI model, to output the CSI corresponding to the first frequency domain resource at t3 through the sixth AI model. The CSI can be CSI bits.

[0143] In some embodiments, the second device determines the AI model based on the first channel information, and takes the second channel information as an input of the AI model to output the CSI corresponding to the t2 moment, which can include: the second device determines a sixth AI model according to the first channel information, and takes the second channel information corresponding to the M antenna ports and the second frequency domain resource at multiple different moments as an input of the sixth AI model, and outputs the CSI corresponding to the N antenna ports and the first frequency domain resource at the t3 moment through the sixth AI model. The CSI can be CSI bits.

[0144] There are various ways for the second device to determine the sixth AI model according to the first channel information, which are not limited in the embodiments of the present application. As an example, the way for the second device to determine the sixth AI model according to the first channel information can refer to the way of determining the second AI model, which will not be described here.

[0145] After the second device determines the sixth AI model, the second device can take the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different moments as an input of the sixth AI model, so as to obtain the output of the sixth AI model. The output of the sixth AI model is the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment.

[0146] In some embodiments, the second device can take the first channel information and the second channel information as an input of the AI model to output the third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the t2 moment, and determine the CSI corresponding to the t2 moment based on the third channel information, which includes: the second device takes the first channel information and the second channel information corresponding to the M antenna ports at multiple different moments as an input of a seventh AI model, outputs the third channel information corresponding to the N antenna ports at the t3 moment through the seventh AI model, and determines the CSI corresponding to the N antenna ports at the t3 moment according to the third channel information, and / or, the second device takes the first channel information and the second channel information corresponding to the second frequency domain resource at multiple different moments as an input of the seventh AI model, outputs the third channel information corresponding to the first frequency domain resource at the t3 moment through the seventh AI model, and determines the CSI corresponding to the first frequency domain resource at the t3 moment according to the third channel information.

[0147] It should be noted that the third channel information is not the direct CSI bits, but the channel information similar to the second channel information, and the difference between the third channel information and the second channel information is that the number of corresponding antenna ports is different. The second channel information corresponds to part of the channel information of the M antenna ports, while the third channel information corresponds to the complete channel information of the N antenna ports.

[0148] There are many methods for the second device to calculate the CSI corresponding to the N antenna ports at time t3 based on the third channel information, which are not limited in the present application. In some embodiments, the method for the second device to calculate the CSI corresponding to the N antenna ports at time t3 based on the third channel information can reuse the method of the related art, that is, the third channel information is equivalent to the channel information measured by the reference signal through the N antenna ports in the related art. Since the scheme of the present application only needs the second reference signal corresponding to the M antenna ports, the overhead of the reference signal can be reduced.

[0149] There are many methods for the second device to calculate the CSI corresponding to the first frequency domain resource at time t3 based on the third channel information, which are not limited in the present application. In some embodiments, the method for the second device to calculate the CSI corresponding to the first frequency domain resource at time t3 based on the third channel information can reuse the method of the related art, that is, the third channel information is equivalent to the channel information measured by the reference signal on the first frequency domain resource in the related art. Since the scheme of the present application only needs the second reference signal occupying the second frequency domain resource, the overhead of the reference signal can be reduced.

[0150] There are many methods for the second device to calculate the CSI corresponding to the N antenna ports at time t3 based on the third channel information, which are not limited in the present application. In some embodiments, the method for the second device to calculate the CSI corresponding to the N antenna ports and the first frequency domain resource at time t3 based on the third channel information can reuse the method of the related art, that is, the third channel information is equivalent to the channel information measured by the reference signal of the N antenna ports and the first frequency domain resource in the related art. Since the scheme of the present application only needs the second reference signal corresponding to the M antenna ports and the second frequency domain resource, the overhead of the reference signal can be reduced.

[0151] In some embodiments, the second device can determine the AI model based on the first channel information, and take the second channel information as the input of the AI model to output the third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the t2 moment, and determine the CSI corresponding to the t2 moment based on the third channel information, including: the second device determines an eighth AI model according to the first channel information, and takes the first channel information and the second channel information corresponding to the M antenna ports at multiple different moments as the input of the eighth AI model, outputs the third channel information corresponding to the N antenna ports at the t3 moment through the eighth AI model, and determines the CSI corresponding to the N antenna ports at the t3 moment according to the third channel information, and / or, the second device determines an eighth AI model according to the first channel information, and takes the first channel information and the second channel information corresponding to the second frequency domain resource at multiple different moments as the input of the eighth AI model, outputs the third channel information corresponding to the first frequency domain resource at the t3 moment through the eighth AI model, and determines the CSI corresponding to the first frequency domain resource at the t3 moment according to the third channel information.

[0152] The method for the second device to determine the eighth AI model is similar to the method for determining the seventh AI model, which will not be described here.

[0153] After determining the CSI, the second device can send the CSI to the first device, or the second device can perform precoding on data according to the CSI and send the precoded data to the first device.

[0154] In some embodiments, after the second device determines the CSI, the second device can directly send the CSI to the first device. The way for the second device to send the CSI is not limited in the embodiments of the present application. In some implementations, the second device can send the CSI to the first device through PUCCH. In another implementation, the second device can send the CSI to the first device through PUSCH. In another implementation, the second device can send the CSI to the first device through the scheduling DCI.

[0155] After receiving the CSI sent by the second device, the first device can determine the precoding matrix used for uplink and / or downlink transmission according to the CSI. In some embodiments, the first device can employ an AI model to recover the precoding matrix, and the AI model employed by the first device can be matched with the AI model used by the second device, facilitating the recovery of the precoding matrix from the CSI information. That is, the first device can employ a double-end model to recover the precoding information. For example, for the second device, the input of the AI model can include the second channel information measured by the M antenna ports on the second frequency domain resource, and can also include the first channel information corresponding to the N antenna ports measured previously, and the output is the CSI bit. For the first device, the input of the AI model is the CSI bit, and the output is the precoding matrix or eigenvector corresponding to the N antenna ports on the first frequency domain resource. In other embodiments, when the CSI includes the PMI, the first device can determine the code word corresponding to the PMI from the codebook according to the PMI and the predefined codebook, thereby obtaining the precoding matrix.

[0156] In other embodiments, after the second device determines the CSI corresponding to the t2 moment and / or the t3 moment, the second device performs precoding on the data according to the CSI, and sends the precoded data to the first device.

[0157] The determination manner of the M antenna ports is not specifically limited in the embodiments of the present application. As an example, the M antenna ports can be predefined by a protocol. As another example, the M antenna ports can be indicated by the first device. As yet another example, the M antenna ports can be determined based on the N antenna ports. As yet another example, the value of M can be indicated by the second device to the first device through UE capability, which is used to indicate the number of partial antenna ports required by the terminal device to support the recovery of a complete channel. In another manner, the terminal device can report the minimum value of M through UE capability, so that the network device can configure an M greater than or equal to the minimum value. The determination manner of the M antenna ports based on the N antenna ports is introduced below.

[0158] In some embodiments, the M antenna ports can have a mapping relationship with the N antenna ports. The mapping relationship includes any one of the following: the M antenna ports are obtained by beamforming the N antenna ports; the M antenna ports are obtained by sampling the N antenna ports in the horizontal dimension; the M antenna ports are obtained by sampling the N antenna ports in the vertical dimension; the M antenna ports are the first M antenna ports of the N antenna ports; the M antenna ports are the last M antenna ports of the N antenna ports; or the M antenna ports are the {n*k+1, n*k+2,.., n*k+M / 2}th ports in the N antenna ports, where k=0, 1, and n=N / 2.

[0159] In some embodiments, the M antenna ports are obtained by beamforming from the N antenna ports. For example, the N antenna ports are beamformed into the M antenna ports. The beamforming weight can be determined by the first device, which can be a fixed value or a non-fixed value.

[0160] In some other embodiments, the M antenna ports are obtained by sampling in the horizontal dimension from the N antenna ports. That is, the first device transmits part of the antenna ports with a certain interval in the horizontal dimension. Referring to FIG. 6, the interval in the horizontal dimension is 1, and the first device can transmit part of the antenna ports with the interval in the horizontal dimension, and the solid line part is the obtained M antenna ports.

[0161] In some other embodiments, the M antenna ports are obtained by sampling in the vertical dimension from the N antenna ports. That is, the first device transmits part of the antenna ports with a certain interval in the vertical dimension. Referring to FIG. 7, the interval in the vertical dimension is 1, and the first device can transmit part of the antenna ports with the interval in the vertical dimension, and the solid line part is the obtained M antenna ports.

[0162] In some other embodiments, the M antenna ports are the first M antenna ports of the N antenna ports. That is, the first device transmits part of the antenna ports with a certain interval in the horizontal and vertical dimensions. Referring to FIG. 8, the interval in the horizontal and vertical dimensions is 1, and the first device can transmit part of the antenna ports with the interval in the horizontal and vertical dimensions, and the solid line part is the obtained M antenna ports.

[0163] In some other embodiments, the M antenna ports are the last M antenna ports of the N antenna ports. Referring to FIG. 9, the solid line part is the obtained M antenna ports.

[0164] In some other embodiments, the M antenna ports are the {n*k+1, n*k+2,.., n*k+M / 2}th ports of the N antenna ports, where k=0, 1, and n=N / 2. Based on this manner, the M antenna ports are actually the first M / 2 groups of dual-polarized antennas of the N antenna ports. Referring to FIG. 10, the solid line part is the obtained M antenna ports.

[0165] In some embodiments, the mapping relationship between the M antenna ports and the N antenna ports can be indicated by the first device to the second device. For example, the first device can send first indication information to the second device, where the first indication information is used to indicate the mapping relationship between the M antenna ports and the N antenna ports. That is, the first indication information can indicate the currently used mapping relationship from a plurality of agreed mapping manners, which can include one or more of the aforementioned mapping manners. Directly indicating the mapping relationship between the M antenna ports and the N antenna ports by the first indication information can improve the reliability of the second device in recovering the CSI of the N antenna ports.

[0166] The second device can determine the CSI corresponding to the N antenna ports at the time t2 or the time t3 according to the mapping relationship. The second device determining the CSI corresponding to the N antenna ports at the time t2 or the time t3 includes: the second device determining an AI model used when calculating the CSI according to the mapping relationship; or the second device taking the mapping relationship as an input parameter of the AI model used when calculating the CSI to output the CSI or output third channel information used to calculate the CSI.

[0167] In some embodiments, the second device can determine the AI model used when calculating the CSI according to the mapping relationship, that is, the first AI model to the eighth AI model mentioned above can be selected from a plurality of candidate models based on the mapping relationship. Different mapping relationships (mapping manners) can correspond to different AI models. For example, each mapping relationship corresponds to an AI model, and the second device can determine the corresponding AI model to output the CSI according to the mapping relationship indicated by the first device.

[0168] In other embodiments, the second device can take the mapping relationship as an input parameter of the AI model used when calculating the CSI to output the CSI or output third channel information used to calculate the CSI. For example, the second device takes the mapping relationship as an input parameter of the AI model to output the third channel information and calculate the CSI based on the third channel information. For another example, the second device takes the mapping relationship as an input parameter of the AI model to directly output the CSI. The mapping relationship can be an input parameter of the first AI model to the eighth AI model mentioned above.

[0169] If the first channel information is outdated, it will cause poor performance of the CSI feedback. In general, the second device can continuously monitor the performance of the CSI, and when the performance of the CSI feedback is monitored to be degraded, the second device needs to update the first channel information.

[0170] There are many ways to update the first channel information, and embodiments of the present application do not make specific limitations on this. In some embodiments, the second device receives a third reference signal sent by the first device; and the second device updates the first channel information based on the third reference signal. That is, the first device can send a third reference signal to the second device, so that the second device can obtain the first channel information based on the third reference signal, thereby replacing the first channel information obtained based on the first reference signal described above.

[0171] Embodiments of the present application do not make specific limitations on the way the first device sends the third reference signal. For example, the first device can periodically send the third reference signal. For another example, the first device can send the third reference signal aperiodically. The first device can send the third reference signal to the second device upon receiving a request message from the second device. The second device can send the request message to the first device upon monitoring that the first channel information is outdated.

[0172] In some embodiments, in the case that the second device needs to update the first channel information, the second device can send second indication information to the first device, for indicating the first device to send a third reference signal, the third reference signal corresponding to N antenna ports and / or occupying the first frequency domain resource. For example, the second device sends second indication information to the first device, requesting the second device to send a third reference signal containing N antenna ports, and after the first device receives the request information, the first device sends the third reference signal, so that the second device can determine the first channel information based on the third reference signal.

[0173] The first reference signal and the second reference signal sent by the first device can be quasi co-located. The second device can assume that the channel large-scale parameters of multiple reference signals are the same as the channel large-scale parameters of the first reference signal, and the channel large-scale parameters can include one or more of the following information: spatial reception parameters, Doppler shift, Doppler spread, delay spread, average delay, etc. Therefore, the second device can receive multiple reference signals (such as the second reference signal) based on the channel large-scale parameters obtained from the first reference signal. The channel characteristics on the antenna port symbols corresponding to the second reference signal can be derived from the antenna port corresponding to the first reference signal, that is, the channel estimation result obtained from the antenna port corresponding to the first reference signal can be used for the antenna port corresponding to the second reference signal.

[0174] In an embodiment, the first device can send, to the second device, third indication information for indicating a quasi co-location relationship between the first reference signal and the second reference signal. The second device can determine, according to the quasi co-location relationship, the first reference signal corresponding to the second reference signal, and determine the CSI at the time t2 and / or the time t3 based on the first reference signal and the second reference signal. For example, the first device can configure a CSI-RS resource of the first reference signal as a quasi co-location signal to a CSI-RS resource of the second reference signal, so that the second device can determine the first reference signal corresponding to the second reference signal according to the CSI-RS resource of the second reference signal. For another example, the first device can configure a SRS resource of the first reference signal as a quasi co-location signal to a SRS resource of the second reference signal, so that the second device can determine the first reference signal corresponding to the second reference signal according to the SRS resource of the second reference signal.

[0175] The first reference signal and the second reference signal can employ different periodic configurations and / or different frequency domain densities when the first device transmits the first reference signal and the second reference signal. In some embodiments, the first reference signal and the second reference signal have different periods. In other embodiments, the first reference signal and the second reference signal have different frequency domain densities.

[0176] The first reference signal and the second reference signal satisfy one or more of the following: the first reference signal and the second reference signal are periodically transmitted reference signals, the period of the first reference signal is greater than the period of the second reference signal; the frequency domain density of the first reference signal is less than the frequency domain density of the second reference signal; the first reference signal is a periodically transmitted reference signal, and the second reference signal is a non-periodic or semi-persistent reference signal; the first reference signal is a non-periodic reference signal, and the second reference signal is a periodic or semi-persistent reference signal.

[0177] In some embodiments, the first reference signal and the second reference signal can be periodically transmitted reference signals, and the period of the first reference signal is greater than the period of the second reference signal. For example, the transmission period of the first reference signal is T1, and the transmission period of the second reference signal is T2, T1>T2. The transmission period of the first reference signal being greater than the transmission period of the second reference signal can reduce the overhead of the first device transmitting the first reference signal, and thus can reduce power consumption.

[0178] In other embodiments, the frequency domain density of the first reference signal is less than the frequency domain density of the second reference signal. For example, the frequency domain density of the first reference signal is P1, and the frequency domain density of the second reference signal is P2, P1<P2. The frequency domain density of the first reference signal being less than the frequency domain density of the second reference signal can reduce the overhead of the first device transmitting the first reference signal, and thus can reduce power consumption.

[0179] In some embodiments, the first reference signal is a periodic reference signal, and the second reference signal is an aperiodic or semi-persistent reference signal. Here, semi-persistent means that the second reference signal is activated and deactivated by a MAC CE and is periodically transmitted during the activation period. For example, the first reference signal can be transmitted periodically, and the second reference signal is transmitted aperiodically. For another example, the first reference signal can be transmitted periodically, and the second reference signal is transmitted semi-persistently. That is, before transmitting the second reference signal, the second reference signal is activated by a MAC CE, and then the second reference signal is periodically transmitted to the second device until the MAC CE triggers deactivation, and the first device stops transmitting the second reference signal to the second device.

[0180] In some embodiments, the first reference signal is an aperiodic reference signal, and the second reference signal is a periodic or semi-persistent reference signal. For example, the first reference signal is transmitted aperiodically, and the second reference signal is transmitted periodically. For another example, the first reference signal is transmitted aperiodically, and the second reference signal is transmitted semi-persistently.

[0181] It should be noted that the AI model in the above can be any neural network model. The AI model can also be replaced by an ML model.

[0182] In an implementation, the training data used for training the model can be obtained based on signals similar to the first reference signal and the second reference signal and corresponding channel measurement processes. To ensure consistency between the training and inference processes, the second device can indicate the configuration (e.g., period size) of the reference signal used in the training data obtaining process to the first device, so that the first device can use similar configurations for the first reference signal and the second reference signal, thereby ensuring the performance of the model inference. For example, the terminal device can indicate the period configuration of the first reference signal used in the training process to the network device.

[0183] The schemes of the embodiments of the present application will be described in detail below in conjunction with six embodiments. It should be noted that the embodiments 1 to 6 are only for the convenience of understanding and the introduction of the present application, and the embodiments of the present application should not be limited thereto.

[0184] Embodiment 1

[0185] Embodiment 1 is a scheme for determining downlink CSI by spatial prediction. In this scenario, the first device is a network device, and the second device is a terminal device.

[0186] The first device transmits the first reference signal at time t1, and the first reference signal contains N antenna ports.

[0187] In an embodiment, the first reference signal is a CSI-RS signal, corresponding to one CSI-RS resource. N can be 32, 64, 128, or 256, etc. Each antenna port corresponds to one antenna radio frequency unit. t1 corresponds to the time domain resource configured by the first device for transmitting the first reference signal.

[0188] The second device obtains the first channel information corresponding to the N antenna ports at t1 based on the first reference signal transmitted by the first device.

[0189] In an embodiment, the first channel information is one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, value of N, LOS / NLOS indication, and multipath parameter of the channel. For example, the first channel information can be a K*N dimensional channel matrix, where K is the number of receiving antennas of the UE. For another example, the first channel information can be an N*N dimensional channel covariance matrix. For another example, the first channel information can be several N*1 dimensional channel eigenvectors. For another example, the first channel information can be an indication of LOS / NLOS, indicating whether the measured channel is a LOS channel. For another example, the first channel information can be a multipath parameter of the measured channel, such as the time delay and power spectrum of the multipath. For another example, the first channel information can further include the value of N based on the aforementioned information.

[0190] The first device transmits a second reference signal at t2, which contains M antenna ports, where M < N, and t2 is after t1.

[0191] In an embodiment, the first device receives the first capability information transmitted by the second device, indicating that the second device has the capability of performing CSI measurement based on partial antenna ports. That is, the capability indicates whether the terminal device can obtain the CSI corresponding to the current N antenna ports based on the reference signal of the N antenna ports previously transmitted by the network device and the reference signal of the M antenna ports currently transmitted. If the terminal device indicates that it has the capability, the reference signal resource overhead can be reduced by the method of the present application; otherwise, the network device needs to continuously transmit the reference signal of the N antenna ports.

[0192] In an embodiment, t2 corresponds to the time domain resource configured by the first device for transmitting the second reference signal, where t2 is after t1, indicating that the time domain resource for transmitting the second reference signal is after the time domain resource for transmitting the first reference signal, i.e., the second device receives the first reference signal first and then receives the second reference signal.

[0193] In an embodiment, the M antenna ports (antennas) for transmitting the second reference signal have a mapping relationship with the N antenna ports (antennas) for transmitting the first reference signal. The mapping relationship can include one of the following mapping relationships.

[0194] The M antenna ports are obtained by beamforming from the N antenna ports. The weight of the beamforming is determined by the first device, and the weight is fixed and unchanged.

[0195] The M antenna ports are obtained by sampling in the horizontal dimension from the N antenna ports. For example, the M antenna ports are obtained by sampling in the horizontal dimension from the N antenna ports, that is, a certain interval is used to transmit part of the antenna ports in the horizontal dimension, as shown in FIG. 6 (solid line part).

[0196] The M antenna ports are obtained by sampling in the vertical dimension from the N antenna ports. For example, the M antenna ports are obtained by sampling in the vertical dimension from the N antenna ports, that is, a certain interval is used to transmit part of the antenna ports in the vertical dimension, as shown in FIG. 7 (solid line part).

[0197] The M antenna ports are obtained by sampling in the vertical and horizontal dimensions from the N antenna ports. For example, the M antenna ports are obtained by sampling in the vertical and horizontal dimensions from the N antenna ports, that is, a certain interval is used to transmit part of the antenna ports in the horizontal and vertical dimensions, as shown in FIG. 8 (solid line part).

[0198] The M antenna ports are the first M antenna ports of the N antenna ports. As shown in FIG. 9 (solid line part).

[0199] The M antenna ports are the last M antenna ports of the N antenna ports; as shown in FIG. 10 (solid line part).

[0200] The M antenna ports are the {n*k+1, n*k+2,.., n*k+M / 2}th ports in the N antenna ports, where k=0, 1, and n=N / 2. Based on this manner, the M antenna ports are actually the first M / 2 groups of dual-polarized antennas in the N antenna ports, as shown in FIG. 11 (solid line part).

[0201] In an embodiment, the first device sends first indication information to the second device, and the first indication information is used to indicate the mapping relationship between the M antenna ports and the N antenna ports. That is, the first indication information can indicate the currently used mapping manner from a plurality of agreed mapping manners, and the plurality of agreed mapping manners can include one or more of the foregoing mapping manners. The first indication information can be used by the second device to determine the CSI corresponding to the N antenna ports at t2 or t3 according to the mapping relationship, and specific reference is made to the description of other steps.

[0202] In an implementation, the first reference signal and the second reference signal correspond to different CSI-RS resources or different sets of CSI-RS resources.

[0203] In an implementation, the first device sends third indication information to the second device, the third indication information being used to indicate the quasi co-location relationship between the first reference signal and the second reference signal. For example, the network device can configure the CSI-RS resource of the first reference signal as a quasi co-location signal to the CSI-RS resource of the second reference signal.

[0204] In an implementation, the first reference signal and the second reference signal are configured with different periodicity and / or frequency domain density. The first reference signal and the second reference signal can satisfy one or more of the following: the first reference signal and the second reference signal are periodically transmitted reference signals, the periodicity of the first reference signal is greater than the periodicity of the second reference signal; the frequency domain density of the first reference signal is less than the frequency domain density of the second reference signal; the first reference signal is a periodically transmitted reference signal, and the second reference signal is a non-periodic or semi-persistent reference signal; the first reference signal is a non-periodic reference signal, and the second reference signal is a periodic or semi-persistent reference signal. Here, semi-persistent means activated and deactivated by MAC CE, and periodically transmitted during the activation period.

[0205] The second device obtains the second channel information corresponding to the M antenna ports at time t2 based on the second reference signal sent by the first device.

[0206] In an implementation, the second device sends the first capability information to the first device, which is used to indicate that the second device has the capability of performing CSI measurement based on part of the antenna ports. That is, the capability indicates whether the terminal device can obtain the CSI corresponding to the N antenna ports based on the reference signal of the N antenna ports previously sent by the network device and the reference signal of the M antenna ports currently sent. If the terminal device indicates that it has the capability, the network device can configure the terminal device to reduce the reference signal resource overhead by the method of the present application; otherwise, the terminal device can only obtain the CSI based on the reference signal of the N antenna ports.

[0207] In an implementation, the first reference signal and the second reference signal are quasi co-located. That is, the second device can assume that the channel large-scale parameters of the second reference signal and the channel large-scale parameters of the first reference signal are the same. The channel large-scale parameters can include spatial reception parameters, Doppler shift, Doppler spread, delay spread, average delay, etc. Therefore, the second device can use the channel large-scale parameters obtained from the first reference signal to receive the second reference signal.

[0208] In an embodiment, the second device receives third indication information sent by the first device, the third indication information being used to indicate a quasi-co-location relationship between the first reference signal and the second reference signal. The second device can determine the first reference signal corresponding to the second reference signal according to the quasi-co-location relationship, and determine the CSI corresponding to the N antenna ports at the time t2 based on the first reference signal and the second reference signal.

[0209] In an embodiment, the second channel information is one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, and value of M. For example, the second channel information can be a channel matrix with a dimension of K*M, where K is the number of receive antennas of the UE; for example, the second channel information can be a channel covariance matrix with a dimension of M*M; for example, the second channel information can be several channel eigenvectors with a dimension of M*1. Further, the second channel information can further include the value of M on the basis of the foregoing information.

[0210] The second device determines the CSI corresponding to the N antenna ports at the time t2 based on the first channel information and the second channel information, and sends the CSI to the first device.

[0211] In an embodiment, the second device can determine the CSI corresponding to the N antenna ports at the time t2 by using one of the following methods.

[0212] The second device takes the first channel information and the second channel information as inputs of the first AI model, and outputs the CSI corresponding to the N antenna ports at the time t2 through the first AI model. The CSI output by the AI model is the CSI bit, which can be fed back to the first device through the uplink control channel.

[0213] The second device determines the second AI model according to the first channel information, takes the second channel information as an input of the second AI model, and outputs the CSI corresponding to the N antenna ports at the time t2 through the second AI model.

[0214] In an embodiment, the second device can determine the second AI model according to the first channel information by using two methods.

[0215] Method 1: Different values of the first channel information (such as LOS or NLOS, value of N, multipath parameter, etc.) correspond to different AI models, and the second device determines the second AI model from a plurality of candidate models according to the value of the first channel information.

[0216] Method 2: The second device takes the first channel information as an input of another AI model, which is used to determine the second AI model. The function of this AI model is to determine the second AI model from a plurality of candidate models.

[0217] The CSI output by the AI model is CSI bits, which can be fed back to the first device through an uplink control channel.

[0218] The second device takes the first channel information and the second channel information as inputs of a third AI model, and outputs third channel information corresponding to N antenna ports at time t2 through the third AI model; and calculates CSI corresponding to the N antenna ports according to the third channel information.

[0219] In an implementation, the third channel information is not CSI bits directly, but channel information similar to the second channel information, and the difference between the third channel information and the second channel information is that the number of corresponding antenna ports is different. The second channel information is partial channel information corresponding to M antenna ports, and the third channel information is complete channel information corresponding to N antenna ports.

[0220] In an implementation, the process of calculating CSI corresponding to N antenna ports based on the third channel information can reuse the method of related technologies. That is, the third channel information is equivalent to the channel information measured by the reference signal of N antenna ports in related technologies, but only M antenna ports of the second reference signal are needed, thereby reducing the overhead of the reference signal.

[0221] The second device determines a fourth AI model according to the first channel information, takes the second channel information as an input of the fourth AI model, and outputs third channel information corresponding to N antenna ports at time t2 through the fourth AI model; and calculates CSI corresponding to the N antenna ports according to the third channel information. The method of determining the fourth AI model is similar to the method of determining the second AI model, which will not be described here.

[0222] In an implementation, the third channel information is one or more of the following information: a channel matrix, a channel covariance matrix, a channel eigenvector, and a value of M. For example, the second channel information can be a channel matrix of K*M dimensions, where K is the number of receive antennas of the UE; for example, the second channel information can be a channel covariance matrix of M*M dimensions; for example, the second channel information can be several channel eigenvectors of M*1 dimensions. Further, the second channel information can further contain the value of M on the basis of the foregoing information. In an implementation, the third channel information and the second channel information are channel information of the same type, for example, both are channel eigenvectors.

[0223] In an implementation, the CSI at least contains precoding matrix indicator (PMI) information. The PMI information can be used by the first device to obtain a precoding matrix of downlink transmission. Further, the CSI can also contain RI, CQI, and other information.

[0224] In an embodiment, the second device can receive the first indication information sent by the first device, the first indication information being used to indicate a mapping relationship between the M antenna ports and the N antenna ports; and the second device determines the CSI corresponding to the N antenna ports at the time t2 according to the mapping relationship.

[0225] In an embodiment, the second device determines an AI model used for calculating the CSI according to the mapping relationship. That is, the first AI model, the second AI model, the third AI model and the fourth AI model mentioned above can be selected from a plurality of candidate models based on the mapping relationship. Different mapping relationships (mapping manners) can correspond to different AI models, for example, each mapping relationship corresponds to an AI model, and the second device determines the corresponding AI model for outputting the CSI according to the mapping relationship indicated by the first device.

[0226] In another embodiment, the second device takes the mapping relationship as an input parameter of the AI model, so as to output the third channel information, and calculates the CSI based on the third channel information. Alternatively, the second device takes the mapping relationship as an input parameter of the AI model, so as to directly output the CSI. For example, the mapping relationship can be taken as an input parameter of the first AI model, the second AI model, the third AI model and the fourth AI model mentioned above.

[0227] In an embodiment, the second device can use the following method to train the AI model (such as the first to fourth AI models) used to obtain the CSI.

[0228] The second device can train the AI model based on the training data sets collected respectively under different antenna configurations, different antenna mapping manners and different channel environments, so that the trained AI model can be applicable to different antenna configurations, antenna mapping manners and channel environments. For example, the AI model can identify the current antenna configuration, antenna mapping manner and channel environment through the first channel information, and then adopt the matching model parameters.

[0229] The second device can train the AI model respectively for different antenna configurations, and different AI models correspond to different antenna configurations, so that the AI model adopted can be determined according to the antenna configuration determined by the first channel information.

[0230] The second device can train the AI model respectively for different antenna mapping manners, and different AI models correspond to different antenna mapping manners, so that the AI model adopted can be determined according to the antenna mapping manner determined by the first channel information and the second channel information, or according to the antenna mapping manner indicated by the first device.

[0231] The second device can train AI models for different channel environments respectively, and different AI models correspond to different channel environments, so that the AI model used can be determined according to the channel environment parameters determined by the first channel information.

[0232] In an embodiment, the second device can feed back the CSI to the first device through PUCCH or PUSCH.

[0233] The first device receives the CSI indicated by the second device.

[0234] In an embodiment, the first device can determine the precoding matrix used for downlink transmission according to the CSI.

[0235] In an embodiment, the first device can use an AI model matched with the AI model used by the second device to recover the precoding matrix from the CSI information. That is, the recovery of the precoding information can be performed in a double-end model manner.

[0236] In another embodiment, when the CSI contains PMI, the first device can determine the code word corresponding to the PMI from the codebook according to the PMI and the predefined codebook, so as to obtain the precoding matrix.

[0237] The terminal device can obtain the information such as antenna configuration, antenna mapping mode and channel environment based on the complete first channel information, so that the recovery process of the partial (second) channel information to the complete channel information can be completed without the configuration and assistance of the network device. In this case, the same model can be used for different cells and different terminal devices, and the model interaction between the terminal device and the network device can be avoided, thereby saving a large amount of signaling and time related to model management. At the same time, since the AI model used can well match the information such as antenna configuration, antenna mapping mode and channel environment, the reliability of the CSI measurement can be significantly improved. Since the network device only needs to send the first reference signal at a very low frequency (or even only once), the CSI acquisition can be based on the second reference signal, thereby significantly reducing the overhead of the reference signal and the complexity of the terminal.

[0238] Embodiment 2

[0239] Embodiment 2 is a scheme for determining downlink CSI through spatial prediction and time prediction. In this scenario, the first device is a network device, and the second device is a terminal device.

[0240] The first device sends a first reference signal at t1, and the first reference signal contains N antenna ports.

[0241] In an embodiment, the first reference signal is a CSI-RS signal, which corresponds to one CSI-RS resource.

[0242] In an embodiment, N can be 32, 64, 128, or 256, etc. In an embodiment, each antenna port corresponds to one antenna radio frequency unit.

[0243] In an embodiment, t1 corresponds to the time domain resource configured by the first device for sending the first reference signal.

[0244] The second device obtains the first channel information corresponding to the N antenna ports at t1 based on the first reference signal sent by the first device.

[0245] The first device sends a plurality of reference signals between t1 and t3, the plurality of reference signals including the second reference signal sent at t2, and the plurality of reference signals each including M antenna ports, where M < N, and t3 is after t1 and t2.

[0246] In an embodiment, the first device receives the first capability information sent by the second device, which is used to indicate that the second device has the capability of CSI prediction based on partial antenna ports. That is, the capability indicates whether the terminal device can obtain the CSI corresponding to the N antenna ports at a future time based on the reference signals of the N antenna ports previously sent by the network device and the reference signals of the M antenna ports recently sent. If the terminal device has the capability, the network device can reduce the reference signal resource overhead and configure the terminal device to perform CSI prediction by the method of the present application; otherwise, the network device needs to continuously send the reference signals of the N antenna ports for obtaining the current CSI.

[0247] In an embodiment, t2 corresponds to the time domain resource configured by the first device for sending the second reference signal. Wherein, the plurality of reference signals are sent between t1 and t3, which means that the time domain resources of the plurality of reference signals are between t1 and t3, i.e. the terminal device needs to receive the plurality of reference signals after t1 and before t3.

[0248] In an embodiment, the M antenna ports (antennas) used for sending the plurality of reference signals have a certain mapping relationship with the N antenna ports (antennas) used for sending the first reference signal. The specific mapping relationship can refer to the description in Embodiment 1. Further, the first device uses the same M antenna ports to send the plurality of reference signals, i.e. the plurality of reference signals are quasi co-located (QCL).

[0249] In an embodiment, the first device sends the first indication information to the second device, which is used to indicate the mapping relationship between the M antenna ports and the N antenna ports.

[0250] In an embodiment, the first reference signal and the multiple reference signals employ different sets of CSI-RS resources, and the multiple reference signals employ different CSI-RS resources in the same set of CSI-RS resources.

[0251] In an embodiment, the first reference signal and the multiple reference signals (including the second reference signal) employ different periodicity configurations and / or frequency domain densities. Specifically, one or more of the following can be employed:

[0252] The first reference signal and the multiple reference signals are periodically transmitted reference signals, the periodicity of the first reference signal is greater than the periodicity of the multiple reference signals, and the periodicity of the multiple reference signals is the same.

[0253] The frequency domain density of the first reference signal is less than that of the multiple reference signals, and the frequency domain densities of the multiple reference signals are the same.

[0254] The first reference signal is a periodically transmitted reference signal, the multiple reference signals are non-periodic or semi-persistent reference signals, and the multiple reference signals employ the same signaling trigger.

[0255] The first reference signal is a non-periodic reference signal, the multiple reference signals are periodic or semi-persistent reference signals, and the periodicity of the multiple reference signals is the same.

[0256] The second device obtains second channel information corresponding to the M antenna ports at multiple time instants based on the multiple reference signals transmitted by the first device between time instants t1 and t3, and the multiple reference signals include the second reference signal transmitted at time instant t2.

[0257] In an embodiment, the first capability information transmitted by the second device to the first device is used to indicate whether the second device has the capability of performing CSI prediction based on partial antenna ports. That is, the capability indicates whether the terminal device can obtain the CSI corresponding to the N antenna ports at a future time instant based on the reference signals of the N antenna ports previously transmitted by the network device and the reference signals of the M antenna ports recently transmitted. If the terminal device has the capability, the reference signal resource overhead can be reduced and CSI prediction can be performed by the method of the present application; otherwise, the terminal device needs to continuously measure the reference signals of the N antenna ports for obtaining the current CSI.

[0258] In an embodiment, the first reference signal and the multiple reference signals are quasi co-located. That is, the second device can assume that the channel large scale parameters of the multiple reference signals are the same as the channel large scale parameters of the first reference signal. The channel large scale parameters here can include spatial receive parameters, Doppler shift, Doppler spread, delay spread, average delay, etc. Therefore, the second device can use the channel large scale parameters obtained from the first reference signal for receiving the multiple reference signals.

[0259] In an embodiment, the second device receives third indication information sent by the first device, the third indication information being used to indicate the quasi co-location relationship between the first reference signal and the multiple reference signals. The second device can determine the first reference signal corresponding to the multiple reference signals according to the quasi co-location relationship, so as to determine the CSI at the time t3 based on the first reference signal and the multiple reference signals jointly.

[0260] In an embodiment, the first reference signal and the multiple reference signals of the M antenna ports use different CSI-RS resource sets, and the multiple reference signals use different CSI-RS resources in the same CSI-RS resource set for transmission.

[0261] In an embodiment, the first reference signal and the multiple reference signals (including the second reference signal) use different periodic configurations and / or frequency domain densities. For details, refer to the description in Embodiment 1.

[0262] The second device determines the CSI corresponding to the N antenna ports at the time t3 according to the first channel information and the second channel information at multiple times and sends it to the first device.

[0263] In an embodiment, the second device can determine the CSI corresponding to the N antenna ports at the time t3 by using one of the following methods.

[0264] The second device takes the first channel information and the second channel information at multiple times as the input of the fifth AI model, and outputs the CSI corresponding to the N antenna ports at the time t3 through the fifth AI model. The CSI output by the AI model is the CSI bit, which can be fed back to the first device through the uplink control channel.

[0265] The second device determines the sixth AI model according to the first channel information, takes the second channel information at multiple times as the input of the sixth AI model, and outputs the CSI corresponding to the N antenna ports at the time t3 through the sixth AI model. For details of the method of determining the AI model according to the first channel information, refer to the description in Embodiment 1.

[0266] The second device takes the first channel information and the second channel information at multiple time points as input of the seventh AI model, and outputs third channel information corresponding to the N antenna ports at the t3 time point through the seventh AI model; and calculates the CSI corresponding to the N antenna ports at the t3 time point according to the third channel information.

[0267] The second device determines the eighth AI model according to the first channel information, takes the second channel information at multiple time points as input of the eighth AI model, and outputs the third channel information corresponding to the N antenna ports at the t3 time point through the eighth AI model; and calculates the CSI corresponding to the N antenna ports at the t3 time point according to the third channel information. The method for determining the AI model according to the first channel information can refer to the description in Embodiment 1.

[0268] In an implementation manner, the second channel information at multiple time points includes the second channel information at the t2 time point obtained based on the second reference signal.

[0269] In an implementation manner, the CSI at least includes precoding matrix indicator (PMI) information. The PMI information can be used by the first device to obtain a precoding matrix of the downlink transmission. Further, the CSI can also include RI, CQI and other information.

[0270] In an implementation manner, the second device can receive the first indication information sent by the first device, and the first indication information is used to indicate the mapping relationship between the M antenna ports and the N antenna ports; and the second device determines the specific description of the CSI mapping relationship corresponding to the N antenna ports at the t3 time point according to the mapping relationship, which can refer to Embodiment 1.

[0271] In an implementation manner, the second device determines the AI model used for calculating the CSI according to the mapping relationship. That is, the fifth AI model, the sixth AI model, the seventh AI model and the eighth AI model mentioned above can be selected from a plurality of candidate models based on the mapping relationship. Different mapping relationships (mapping modes) can correspond to different AI models, for example, each mapping relationship corresponds to an AI model, and the second device determines the corresponding AI model according to the mapping relationship indicated by the first device to output the CSI.

[0272] In another implementation manner, the second device takes the mapping relationship as an input parameter of the AI model, so as to output the third channel information and calculate the CSI based on the third channel information. Or, the second device takes the mapping relationship as an input parameter of the AI model, so as to directly output the CSI. For example, the mapping relationship can be taken as an input parameter of the fifth AI model, the sixth AI model, the seventh AI model and the eighth AI model mentioned above.

[0273] In an implementation manner, the method for training the AI model used to obtain the CSI by the second device can refer to the description in Embodiment 1.

[0274] In an embodiment, the second device sends second indication information to the first device, and the second indication information is used to request the first device to send a third reference signal containing N antenna ports. For example, when the terminal device needs to update the first channel information, the terminal device can send indication information to the network device to request the network device to send a third reference signal of complete antenna ports, so as to measure and update the first channel information. Wherein, the terminal device can continuously perform performance monitoring of the CSI, and if the performance of the CSI feedback is not good due to the out-of-date first channel information, the first channel information can be updated through the second indication information. Further, the second device can receive the third reference signal sent by the first device, and obtain the first channel information by using the third reference signal to replace the first channel information obtained by using the first reference signal.

[0275] The first device receives the CSI indicated by the second device.

[0276] In an embodiment, the first device can determine the precoding matrix used for downlink transmission at t3 according to the CSI. The specific method can refer to the description in embodiment 1.

[0277] In an embodiment, the first device receives the second indication information sent by the second device, and the second indication information is used to request the second device to send a third reference signal containing N antenna ports. After receiving the request information, the first device can configure and send the third reference signal in a non-periodic transmission mode.

[0278] The terminal device can obtain the antenna configuration, antenna mapping mode, channel environment and other information based on the complete first channel information, so as to complete the recovery process of the partial (second) channel information at multiple time points in the past to the complete channel information in the future without the configuration and assistance of the network device. In this case, the same model can be used for different cells and different terminal devices, and the model interaction between the terminal device and the network device can be avoided, thereby saving a large amount of signaling and time related to model management. At the same time, since the AI model adopted can well match the current antenna configuration, antenna mapping mode, channel environment and other information, the reliability of the CSI measurement can be obviously improved. Since the network device only needs to send the first reference signal at a very low frequency (or even only once), the CSI acquisition can be based on the second channel information, thereby obviously reducing the overhead of the reference signal and the complexity of the terminal.

[0279] Embodiment 3

[0280] Embodiment 3 is a scheme for determining uplink CSI by spatial prediction. In this scenario, the first device is a terminal device, and the second device is a network device.

[0281] The second device configures the first device to transmit the first reference signal at time t1 and the second reference signal at time t2, where time t2 is after time t1.

[0282] The first device transmits the first reference signal at time t1, and the first reference signal contains N antenna ports.

[0283] In an implementation, the first reference signal is an SRS signal, corresponding to one or more SRS resources. For example, the first reference signal is a reference signal transmitted through N antenna ports of one SRS resource. For another example, the first reference signal is transmitted through multiple SRS resources, the number of ports of different SRS resources is the same, and the ports of these SRS resources together can obtain a reference signal of N antenna ports, for example, a reference signal of N=8 antenna ports is obtained through 4 SRS resources of 2 antenna ports. These SRS resources can be transmitted in one time slot or adjacent time slots, and time t1 can correspond to one SRS transmission window.

[0284] In an implementation, N can be 8, 16, or 32, etc. In an implementation, each antenna port corresponds to one antenna radio frequency unit.

[0285] In an implementation, time t1 corresponds to a time domain resource configured by the second device for transmitting the first reference signal.

[0286] The second device obtains first channel information corresponding to N antenna ports at time t1 based on the first reference signal transmitted by the first device.

[0287] In an implementation, the first channel information is one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, value of N, LOS / NLOS indication, and channel multipath parameter.

[0288] The first device transmits the second reference signal at time t2, and the second reference signal contains M antenna ports, where M<N.

[0289] In an implementation, time t2 corresponds to a time domain resource configured by the second device for transmitting the second reference signal. Time t2 is after time t1, indicating that the time domain resource for transmitting the second reference signal is after the time domain resource for transmitting the first reference signal, i.e., the second device receives the first reference signal first and then receives the second reference signal.

[0290] In an implementation, the M antenna ports (antennas) used for transmitting the second reference signal have a certain mapping relationship with the N antenna ports (antennas) used for transmitting the first reference signal. The specific mapping relationship can refer to the description in Embodiment 1.

[0291] In an embodiment, the first device sends first indication information to the second device, the first indication information being used to indicate a mapping relationship between the M antenna ports and the N antenna ports. That is, the first indication information can indicate a currently used mapping manner from a plurality of agreed mapping manners, the plurality of agreed mapping manners can include one or more of the foregoing mapping manners. The indication information can be used for the second device to determine the CSI corresponding to the N antenna ports at the time t2 or t3 according to the mapping relationship, and specific descriptions can be referred to descriptions of other steps.

[0292] In an embodiment, the first reference signal and the second reference signal correspond to different SRS resources or different SRS resource sets.

[0293] In an embodiment, the first reference signal and the second reference signal adopt different periodic configurations and / or frequency domain densities. Specific manners can be referred to descriptions in Embodiment 1.

[0294] The second device obtains the second channel information corresponding to the M antenna ports at the time t2 based on the second reference signal sent by the first device.

[0295] In an embodiment, the second channel information is one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, and value of M.

[0296] The second device determines the CSI corresponding to the N antenna ports at the time t2 according to the first channel information and the second channel information.

[0297] In an embodiment, specific methods can be referred to descriptions in Embodiment 1.

[0298] In an embodiment, the CSI at least includes precoding matrix indication (PMI) information. The PMI information can be used for the first device to obtain a precoding matrix of uplink transmission. Further, the CSI can also include other information such as transmission rank (RI) and MCS.

[0299] In an embodiment, the second device can receive the first indication information sent by the first device, the first indication information being used to indicate a mapping relationship between the M antenna ports and the N antenna ports; and the second device determines the CSI corresponding to the N antenna ports at the time t2 according to the mapping relationship.

[0300] The second device indicates the CSI to the first device.

[0301] In an embodiment, the second device can indicate the CSI information to the first device through a scheduled DCI.

[0302] In an embodiment, the first device determines a precoding matrix according to the CSI indicated by the second device.

[0303] In an embodiment, the first device can determine a precoding matrix for the uplink transmission according to the CSI. For example, when the CSI contains the PMI, the first device can determine the precoding matrix according to the PMI and a predefined codebook, and determine a code word corresponding to the PMI from the codebook, thereby obtaining the precoding matrix.

[0304] The network device can obtain the antenna configuration, antenna mapping mode, channel environment and the like based on the complete first channel information, thereby completing the recovery process of the partial (second) channel information to the complete channel information of the current measurement. In this case, the same model can be used for different cells and different network devices, and the model interaction between the terminal device and the network device and the model switching on the network device side can be avoided, thereby saving a large amount of signaling and time related to model management. At the same time, since the AI model adopted can well match the current antenna configuration, antenna mapping mode, channel environment and the like, the reliability of the CSI measurement can be significantly improved. Since the terminal device only needs to send the first reference signal at a very low frequency (or even only once), the CSI acquisition can be based on the second reference signal, thereby significantly reducing the overhead of the reference signal and the complexity of the terminal, and also reducing the power consumption of the terminal device.

[0305] Embodiment 4

[0306] Embodiment 4 is a scheme for determining uplink CSI. In this scenario, the first device is a terminal device, and the second device is a network device.

[0307] The second device configures the first device to send the first reference signal at t1 and the second reference signal at t2, where t2 is after t1.

[0308] The first device sends the first reference signal at t1, and the first reference signal contains N antenna ports.

[0309] In an embodiment, the first reference signal is an SRS signal, corresponding to one or more SRS resources. For example, the first reference signal is transmitted through multiple SRS resources for antenna switching, and the number of ports of different SRS resources is the same but corresponds to different antennas. The ports of these SRS resources can be combined to obtain a reference signal of N antenna ports, for example, a reference signal of N=8 antenna ports can be obtained through 4 SRS resources of 2 antenna ports. These SRS resources can be transmitted in one time slot or adjacent time slots, at which time t1 can correspond to an SRS sending window.

[0310] In an embodiment, t1 corresponds to the time domain resource configured by the second device for sending the first reference signal.

[0311] The second device obtains first channel information corresponding to the N antenna ports at time t1 based on the first reference signal sent by the first device.

[0312] In an embodiment, the first channel information is one or more of the following: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of N, a LOS / NLOS indication, a channel multipath parameter.

[0313] The first device sends a second reference signal at time t2, and the second reference signal contains M antenna ports, where M < N.

[0314] In an embodiment, the first reference signal and the second reference signal are both SRSs for downlink CSI acquisition.

[0315] The second device obtains second channel information corresponding to the M antenna ports at time t2 based on the second reference signal sent by the first device.

[0316] In an embodiment, the second channel information is one or more of the following: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of M.

[0317] The second device determines CSI corresponding to the N antenna ports at time t2 based on the first channel information and the second channel information.

[0318] In an embodiment, the specific method can refer to the description in Embodiment 1.

[0319] In an embodiment, the CSI can be a precoding matrix of downlink, a downlink channel matrix, or a downlink eigenvector.

[0320] In an embodiment, the second device can determine (uplink) third channel information corresponding to the N antenna ports at time t2 based on the first channel information and the second channel information according to an AI model, and then obtain the downlink CSI based on the third channel information according to channel reciprocity. In another embodiment, if the channel reciprocity holds, the second device can directly output the downlink CSI based on the first channel information and the second channel information according to the AI model.

[0321] The second device performs precoding on downlink data according to the CSI and sends it to the first device.

[0322] In an embodiment, the CSI is a precoding matrix of downlink, or the second device calculates a precoding matrix of downlink according to the CSI and uses it for precoding of downlink data.

[0323] Embodiment 5

[0324] Embodiment 5 is a scheme for determining downlink CSI by means of frequency domain prediction. In this scenario, the first device is a network device and the second device is a terminal device.

[0325] The first device transmits a first reference signal at time t1, and the first reference signal occupies a first frequency domain resource.

[0326] In an implementation, the first reference signal is a CSI-RS signal, corresponding to one CSI-RS resource.

[0327] In an implementation, the first frequency domain resource can contain a certain bandwidth, or contain several sub-bands, or contain several PRBs, or contain several subcarriers.

[0328] In an implementation, time t1 corresponds to a time domain resource configured by the first device for transmitting the first reference signal.

[0329] The second device obtains first channel information corresponding to the first frequency domain resource at time t1 based on the first reference signal transmitted by the first device.

[0330] In an implementation, the first channel information is one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, value of N, LOS / NLOS indication, and channel multipath parameter.

[0331] In an implementation, the first channel information can be a measured channel multipath parameter, such as multipath delay and power spectrum, etc. For example, the first channel information can be the delay and / or power spectrum of each multipath, or the maximum delay of the channel.

[0332] The first device transmits a second reference signal at time t2, and the second reference signal occupies a second frequency domain resource, wherein the second frequency domain resource is part of the first frequency domain resource, and time t2 is after time t1.

[0333] In an implementation, the first device receives first capability information transmitted by the second device, and the first capability information is used to indicate that the second device has the capability of performing CSI measurement based on partial frequency domain resources. That is, the capability indicates whether the terminal device can obtain the CSI corresponding to the current first (complete) frequency domain resource based on the reference signal of the first (complete) frequency domain resource previously transmitted by the network device and the reference signal of the second (partial) frequency domain resource currently transmitted. If the terminal device indicates that it has this capability, the reference signal resource overhead can be reduced by the method of the present application; otherwise, the network device needs to continuously transmit the reference signal of the first (complete) frequency domain resource.

[0334] In an embodiment, the time t2 corresponds to a time domain resource configured by the first device for transmitting the second reference signal. Wherein, the time t2 is after the time t1, indicating that the time domain resource for transmitting the second reference signal is after the time domain resource for transmitting the first reference signal, i.e., the second device receives the first reference signal first and then receives the second reference signal.

[0335] In an embodiment, the second frequency domain resource can contain a certain bandwidth, or contain a plurality of subbands, or contain a plurality of PRBs, or contain a plurality of subcarriers. Specifically, the first frequency domain resource and the second frequency domain resource correspond to different bandwidths, or the first frequency domain resource and the second frequency domain resource contain different numbers of subbands, or the first frequency domain resource and the second frequency domain resource contain different numbers of subcarriers, or the first frequency domain resource and the second frequency domain resource contain different numbers of PRBs.

[0336] In an embodiment, the first frequency domain resource contains all subbands for which CSI needs to be measured, and the second frequency domain resource contains part of the subbands. For example, the second frequency domain resource is the odd subbands or even subbands in the first frequency domain resource, or the second frequency domain resource is part of the subbands in the first frequency domain resource at intervals of a plurality of subbands.

[0337] In an embodiment, the first frequency domain resource contains all PRBs for which CSI needs to be measured, and the second frequency domain resource contains part of the PRBs. For example, the second frequency domain resource is the odd PRBs or even PRBs in the first frequency domain resource, or the second frequency domain resource is part of the PRBs in the first frequency domain resource at intervals of a plurality of PRBs.

[0338] In an embodiment, the first frequency domain resource contains all subcarriers for which CSI needs to be measured, and the second frequency domain resource contains part of the subcarriers. For example, the second frequency domain resource is the odd subcarriers or even subcarriers in the first frequency domain resource, or the second frequency domain resource is part of the subcarriers in the first frequency domain resource at intervals of a plurality of subcarriers.

[0339] In an embodiment, the first reference signal and the second reference signal correspond to different CSI-RS resources or different sets of CSI-RS resources.

[0340] In an embodiment, the first device sends third indication information to the second device, and the third indication information is used to indicate the quasi-co-location relationship between the first reference signal and the second reference signal. For example, the network device can configure the CSI-RS resource of the first reference signal as a quasi-co-location signal for the CSI-RS resource of the second reference signal.

[0341] In an embodiment, the first reference signal and the second reference signal adopt different periodic configurations and / or frequency domain densities. Specifically, refer to the description in Embodiment 1.

[0342] The second device obtains second channel information corresponding to the second frequency domain resource at time t2 based on the second reference signal sent by the first device.

[0343] In an embodiment, the second device sends first capability information to the first device, and the first capability information is used to indicate that the second device has the capability of performing CSI measurement based on partial frequency domain resources. That is, the capability indicates whether the terminal device can obtain the CSI corresponding to the current complete frequency domain resource based on the reference signal on the complete frequency domain resource previously sent by the network device and the reference signal on the partial frequency domain resource currently sent. If the terminal device indicates that it has the capability, the network device can configure the terminal device to reduce the reference signal resource overhead by the method of the present application; otherwise, the terminal device can only obtain the CSI based on the reference signal on the complete frequency domain resource.

[0344] In an embodiment, the first reference signal and the second reference signal are quasi co-located.

[0345] In an embodiment, the second channel information is one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector. For example, the second channel information can be a K*M-dimensional channel matrix on the second frequency domain resource, where K is the number of receive antennas of the UE; for example, the second channel information can be a M*M-dimensional channel covariance matrix on the second frequency domain resource; for example, the second channel information can be several M*1-dimensional channel eigenvectors on the second frequency domain resource.

[0346] The second device determines the CSI corresponding to the first frequency domain resource at time t2 according to the first channel information and the second channel information and sends it to the first device.

[0347] In an embodiment, the second device can determine the CSI corresponding to the first frequency domain resource at time t2 by one of the following methods.

[0348] The second device takes the first channel information and the second channel information as the input of the first AI model, and outputs the CSI corresponding to the first frequency domain resource at time t2 through the first AI model. The CSI output by the AI model is the CSI bit, which can be fed back to the first device through the uplink control channel.

[0349] The second device determines the second AI model according to the first channel information, takes the second channel information as the input of the second AI model, and outputs the CSI corresponding to the first frequency domain resource at time t2 through the second AI model.

[0350] In an embodiment, the second device can adopt two methods to determine the second AI model according to the first channel information.

[0351] Method 1: Different values of the first channel information (e.g., different maximum multipath delay) correspond to different AI models, and the second device determines the second AI model from the plurality of candidate models according to the value of the first channel information.

[0352] Method 2: The second device takes the first channel information as the input of another AI model, which is used to determine the second AI model, and the function of the AI model is to determine the second AI model from the plurality of candidate models. For example, the measured time delay and power values of multiple paths can be taken as the input of the model.

[0353] The second device takes the first channel information and the second channel information as the input of a third AI model, and outputs the third channel information corresponding to the first frequency domain resource at time t2 through the third AI model; and calculates the CSI corresponding to the first frequency domain resource according to the third channel information.

[0354] In an implementation, the third channel information is not the direct CSI bit, but the channel information similar to the second channel information, and the difference from the second channel information is that the corresponding frequency domain resource is different. The second channel information is part of the channel information corresponding to the second frequency domain resource, and the third channel information is the complete channel information corresponding to the first frequency domain resource.

[0355] In an implementation, the process of calculating the CSI corresponding to the first frequency domain resource based on the third channel information can reuse the method of the related art. That is, the third channel information is equivalent to the channel information measured through the reference signal on the first frequency domain resource in the related art, but only the second reference signal on the second frequency domain resource is needed, thereby reducing the overhead of the reference signal.

[0356] The second device determines a fourth AI model according to the first channel information, takes the second channel information as the input of the fourth AI model, and outputs the third channel information corresponding to the first frequency domain resource at time t2 through the fourth AI model; and calculates the CSI corresponding to the first frequency domain resource according to the third channel information. The specific method of determining the fourth AI model is similar to the method of determining the second AI model described above, which will not be described here.

[0357] In an implementation, the third channel information is one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector. In an implementation, the third channel information and the second channel information are channel information of the same type, for example, both are channel eigenvectors.

[0358] In an implementation, the CSI at least contains precoding matrix indication (PMI) information. The PMI information can be used by the first device to obtain the precoding matrix of the downlink transmission. Further, the CSI can also contain RI, CQI and other information.

[0359] In an embodiment, the second device can train the AI model (e.g., the first to fourth AI models) for obtaining the CSI by using the following method.

[0360] In an embodiment, the second device can train the AI model based on the training data sets collected respectively in different frequency domain resources, so that the trained AI model can be applicable to different frequency domain resource sizes. For example, the AI model can identify the current channel environment (e.g., the multipath delay size) through the first channel information, and then use the matching model parameters.

[0361] In an embodiment, the second device can train the AI model respectively for different channel environments, and different AI models correspond to different channel environments, so that the AI model used can be determined according to the channel environment parameters determined by the first channel information.

[0362] In an embodiment, the second device can feed back the CSI to the first device through the PUCCH or the PUSCH.

[0363] The first device receives the CSI indicated by the second device.

[0364] In an embodiment, the first device can determine the precoding matrix used for downlink transmission according to the CSI.

[0365] In an embodiment, the first device can use an AI model matched with the AI model used by the second device to recover the precoding matrix from the CSI information. That is, the recovery of the precoding information can be performed in a double-end model manner.

[0366] In another embodiment, when the CSI contains the PMI, the first device can determine the code word corresponding to the PMI from the codebook according to the PMI and the predefined codebook, so as to obtain the precoding matrix.

[0367] The terminal device can obtain the multipath channel parameters and other information based on the first channel information corresponding to the complete frequency band, so that the recovery process of the partial frequency band channel information of the current measurement to the complete frequency band channel information can be completed without the configuration and assistance of the network device. In this case, the same model can be used for different cells and different terminal devices, and the model interaction between the terminal device and the network device can be avoided, thereby saving a large amount of signaling and time related to model management. At the same time, since the AI model used can well match the current multipath delay, the reliability of the frequency domain CSI measurement can be significantly improved. Since the network device only needs to send the first reference signal using a small amount of frequency domain resources, the CSI acquisition can be based on the second reference signal, thereby significantly reducing the overhead of the reference signal and the complexity of the terminal.

[0368] Embodiment 6

[0369] Embodiment 6 is a scheme of determining downlink CSI by means of spatial prediction and frequency domain prediction. In this scenario, the first device is a network device, and the second device is a terminal device. In fact, this method can also be used in the uplink case.

[0370] The first device transmits a first reference signal at time t1. The first reference signal corresponds to N antenna ports and occupies a first frequency domain resource.

[0371] In an embodiment, the first reference signal is a CSI-RS signal, corresponding to one CSI-RS resource.

[0372] In an embodiment, the value of N can be 32, 64, 128, 256, etc. In an embodiment, each antenna port corresponds to one antenna radio frequency unit.

[0373] In an embodiment, the first frequency domain resource can contain a certain bandwidth, or contain several sub-bands, or contain several PRBs, or contain several subcarriers.

[0374] In an embodiment, time t1 corresponds to the time domain resource configured by the first device for transmitting the first reference signal.

[0375] The second device obtains first channel information corresponding to the N antenna ports on the first frequency domain resource at time t1 based on the first reference signal transmitted by the first device.

[0376] In an embodiment, the first channel information is one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, value of N, LOS / NLOS indication, channel multipath parameter.

[0377] In an embodiment, the first channel information can be a K*N-dimensional channel matrix on the first frequency domain resource, where K is the number of receive antennas of the UE; for example, the first channel information can be an N*N-dimensional channel covariance matrix on the first frequency domain resource; for example, the first channel information can be several N*1-dimensional channel eigenvectors on the first frequency domain resource; for example, the first channel information can be an indication of LOS / NLOS, indicating whether the current measured channel is a LOS channel; for example, the first channel information can be the multipath parameters of the channel measured on the first frequency domain resource, such as the time delay and power spectrum of the multipath, the maximum time delay, etc. Further, the first channel information can further contain the value of N on the basis of the foregoing information.

[0378] The first device transmits the second reference signal at a time t2, the second reference signal corresponds to M antenna ports and occupies second frequency domain resources, M < N, the second frequency domain resources are part of the first frequency domain resources, and the time t2 is after the time t1.

[0379] In an embodiment, the first device receives first capability information transmitted by the second device, and the first capability information is used to indicate that the second device has the capability of performing CSI measurement based on partial antenna ports and partial frequency domain resources. That is, the capability indicates whether the terminal device can obtain the CSI corresponding to N antenna ports on the current complete frequency domain resources based on the reference signal of N antenna ports on the complete frequency domain resources previously transmitted by the network device and the reference signal of M antenna ports on the partial frequency domain resources currently transmitted. If the terminal device indicates that it has the capability, the reference signal resource overhead can be reduced by the method of the present application; otherwise, the network device needs to continuously transmit the reference signal of N antenna ports on the complete frequency domain resources.

[0380] In an embodiment, the second frequency domain resources can include a certain bandwidth, or include a plurality of subbands, or include a plurality of PRBs, or include a plurality of subcarriers. Specifically, the first frequency domain resources and the second frequency domain resources correspond to different bandwidths, or the first frequency domain resources and the second frequency domain resources include different numbers of subbands, or the first frequency domain resources and the second frequency domain resources include different numbers of subcarriers, or the first frequency domain resources and the second frequency domain resources include different numbers of PRBs.

[0381] In an embodiment, the M antenna ports (antennas) used to transmit the second reference signal have a certain mapping relationship with the N antenna ports (antennas) used to transmit the first reference signal. Specifically, refer to the description of embodiment 1.

[0382] In an embodiment, the first device transmits first indication information to the second device, and the first indication information is used to indicate the mapping relationship between the M antenna ports and the N antenna ports. Specifically, refer to the description of embodiment 1.

[0383] In an embodiment, the first reference signal and the second reference signal correspond to different CSI-RS resources or different sets of CSI-RS resources.

[0384] In an embodiment, the first device transmits third indication information to the second device, and the third indication information is used to indicate the quasi-co-location relationship between the first reference signal and the second reference signal. For example, the network device can configure the CSI-RS resource of the first reference signal as a quasi-co-location signal for the CSI-RS resource of the second reference signal.

[0385] In an embodiment, the first reference signal and the second reference signal are configured with different periodicity and / or frequency domain density. For example, the first reference signal and the second reference signal can be configured with different periodicity and / or frequency domain density as described in Embodiment 1.

[0386] The second device obtains the second channel information corresponding to the M antenna ports on the second frequency domain resource at time t2 based on the second reference signal sent by the first device.

[0387] In an embodiment, the second device sends the first capability information to the first device, and the first capability information is used to indicate that the second device has the capability of performing CSI measurement based on the partial frequency domain resource and the partial antenna ports. That is, the capability indicates whether the terminal device can obtain the CSI corresponding to the N antenna ports on the current complete frequency domain resource based on the reference signal of the N antenna ports on the complete frequency domain resource sent by the network device before and the reference signal of the M antenna ports on the partial frequency domain resource sent by the network device currently. If the terminal device indicates that it has the capability, the network device can configure the terminal device to reduce the reference signal resource overhead by the method of the present application; otherwise, the terminal device can only obtain the CSI based on the reference signal of the N antenna ports on the complete frequency domain resource.

[0388] In an embodiment, the first reference signal and the second reference signal are quasi co-located.

[0389] In an embodiment, the second channel information is one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, and the value of M. For example, the second channel information can be a K*M dimensional channel matrix on the second frequency domain resource, where K is the number of receive antennas of the UE; for example, the second channel information can be a M*M dimensional channel covariance matrix on the second frequency domain resource; for example, the second channel information can be a plurality of M*1 dimensional channel eigenvectors on the second frequency domain resource; for example, the second channel information can be an indication of LOS / NLOS, which is used to indicate whether the channel measured currently is a LOS channel; for example, the second channel information can be the multipath parameters of the channel measured on the first frequency domain resource, such as the delay and power spectrum of the multipath, the maximum delay, etc. Further, the first channel information can further include the value of M on the basis of the foregoing information.

[0390] The second device determines the CSI corresponding to the N antenna ports on the first frequency domain resource at time t2 based on the first channel information and the second channel information, and sends the CSI to the first device.

[0391] In an embodiment, the second device can determine the CSI corresponding to the N antenna ports on the first frequency domain resource at time t2 by one of the following methods.

[0392] The second device takes the first channel information and the second channel information as inputs of the first AI model, and outputs the CSI corresponding to the N antenna ports on the first frequency domain resource at the time t2 through the first AI model. The CSI output by the AI model is the CSI bit, which can be fed back to the first device through the uplink control channel.

[0393] The second device determines a second AI model according to the first channel information, takes the second channel information as an input of the second AI model, and outputs the CSI corresponding to the N antenna ports on the first frequency domain resource at the time t2 through the second AI model.

[0394] The second device takes the first channel information and the second channel information as inputs of the third AI model, and outputs the third channel information corresponding to the N antenna ports on the first frequency domain resource at the time t2 through the third AI model; and calculates the CSI corresponding to the N antenna ports according to the third channel information.

[0395] In an implementation, the third channel information is not the CSI bit directly, but the channel information similar to the second channel information, and the difference from the second channel information is that the corresponding frequency domain resource and the number of antenna ports are different. The second channel information is the partial channel information corresponding to the M antenna ports on the second frequency domain resource, and the third channel information is the complete channel information corresponding to the N antenna ports on the first frequency domain resource.

[0396] In an implementation, the process of calculating the CSI corresponding to the N antenna ports on the first frequency domain resource based on the third channel information can reuse the method of the related technology. That is, the third channel information is equivalent to the channel information measured by the reference signal of the N antenna ports on the first frequency domain resource in the related technology, but only the second reference signal of the M antenna ports on the second frequency domain resource is needed, thereby reducing the overhead of the reference signal.

[0397] The second device determines a fourth AI model according to the first channel information, takes the second channel information as an input of the fourth AI model, and outputs the third channel information corresponding to the N antenna ports on the first frequency domain resource at the time t2 through the fourth AI model; and calculates the CSI corresponding to the N antenna ports on the first frequency domain resource according to the third channel information. The method of determining the fourth AI model is similar to the method of determining the second AI model, which will not be described here.

[0398] In an implementation, the third channel information is one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, and value of M.

[0399] In an implementation, the CSI at least contains precoding matrix indication (PMI) information. The PMI information can be used by the first device to obtain the precoding matrix of the downlink transmission. Further, the CSI can also contain RI, CQI and other information.

[0400] In an embodiment, the second device can receive the first indication information sent by the first device, the first indication information being used to indicate a mapping relationship between the M antenna ports and the N antenna ports; the second device determines the CSI corresponding to the N antenna ports on the first frequency domain resource at the time t2 according to the mapping relationship. For details, refer to the description of embodiment 1.

[0401] In an embodiment, the method for the second device to train the AI model (such as the first to fourth AI models) can refer to the description of embodiment 1.

[0402] In an embodiment, the second device can feed back the CSI to the first device through PUCCH or PUSCH.

[0403] The first device receives the CSI indicated by the second device.

[0404] In an embodiment, the first device can determine the precoding matrix used for downlink transmission according to the CSI.

[0405] In an embodiment, the first device can use an AI model matched with the AI model used by the second device to recover the precoding matrix or the eigenvector from the CSI information. For example, in the second device, the input of the AI model contains the first channel information measured by the M (partial) antenna ports on the second (partial) frequency domain resource (and can also contain the first channel information of the N antenna ports measured before), and the output is the CSI bit; in the first device, the input of the corresponding AI model is the CSI bit, and the output is the precoding matrix or the eigenvector corresponding to the N (complete) antenna ports on the first (complete) frequency domain resource. This method needs the joint training of the models on both sides.

[0406] The terminal device can obtain the antenna configuration, the antenna mapping mode, the multipath delay, and the channel environment information based on the complete first channel information, so as to complete the recovery process of the current measured partial (second) channel information to the complete channel information without the configuration and assistance of the network device. In this case, the same model can be used for different cells and different terminal devices, and the model interaction between the terminal device and the network device can be avoided, thereby saving a large amount of signaling and time related to model management. At the same time, since the AI model used can well match the current antenna configuration, antenna mapping mode, and channel environment, multipath delay, and other information, the reliability of the multi-antenna and large-bandwidth CSI measurement can be significantly improved. Since the network device only needs to send the first reference signal at a very low frequency and with very few frequency domain resources (and can even be sent only once), the CSI acquisition can be based on the second reference signal, thereby significantly reducing the overhead of the reference signal and the complexity of the terminal.

[0407] The method embodiments of the present application are described in detail above in combination with FIG. 1 to FIG. 11, and the device embodiments of the present application are described in detail below in combination with FIG. 12 to FIG. 14. It should be understood that the description of the method embodiments and the description of the device embodiments correspond to each other, and thus, the parts not described in detail can be referred to the foregoing method embodiments.

[0408] FIG. 12 is a schematic block diagram of a communication device according to an embodiment of the present application. The communication device 1200 shown in FIG. 12 can be any one of the first devices described above. The first communication device can include a sending unit 1210.

[0409] The sending unit 1210 is configured to send a first reference signal at time t1, the first reference signal corresponding to N antenna ports and / or occupying a first frequency domain resource; the sending unit is also configured to send a second reference signal at time t2, the second reference signal corresponding to M antenna ports and / or occupying a second frequency domain resource, where M < N, and time t2 is after time t1; the receiving unit 1220 is configured to receive CSI sent by the second device, or the receiving unit 1220 is configured to receive precoded data sent by the second device, the precoded data being data precoded based on the CSI, the CSI corresponding to N antenna ports and / or the first frequency domain resource at time t2 or time t3, the CSI being measured based on the first reference signal and the second reference signal, where time t3 is after time t2.

[0410] In some embodiments, the CSI is measured based on the first reference signal and the second reference signal, including: the CSI is determined based on first channel information corresponding to N antenna ports and / or the first frequency domain resource and second channel information corresponding to M antenna ports and / or the second frequency domain resource, where the first channel information is measured based on the first reference signal, and the second channel information is measured based on the second reference signal.

[0411] In some embodiments, the M antenna ports used for sending the second reference signal have a mapping relationship with the N antenna ports used for sending the first reference signal as follows: the M antenna ports are obtained by beamforming the N antenna ports; or the M antenna ports are obtained by sampling the N antenna ports in the horizontal dimension; or the M antenna ports are obtained by sampling the N antenna ports in the vertical dimension; or the M antenna ports are the first M antenna ports of the N antenna ports; or the M antenna ports are the last M antenna ports of the N antenna ports; or the M antenna ports are the {n*k+1, n*k+2,.., n*k+M / 2}th ports in the N antenna ports, where k = 0, 1, and n = N / 2.

[0412] In some embodiments, the sending unit is further configured to send, to the second device, first indication information used to indicate a mapping relationship between the M antenna ports and the N antenna ports, the first indication information being used by the second device to determine the CSI corresponding to the N antenna ports at the time t2 or t3 according to the mapping relationship.

[0413] In some embodiments, the mapping relationship is used by the second device to determine an AI model used when calculating the CSI; or the mapping relationship is used as an input parameter of the AI model used when calculating the CSI to output the CSI or output third channel information used to calculate the CSI.

[0414] In some embodiments, the receiving unit is further configured to receive second indication information sent by the second device, the second indication information being used to indicate that the first device sends a third reference signal, the third reference signal corresponding to the N antenna ports and / or occupying the first frequency domain resource; and the sending unit is further configured to send the third reference signal.

[0415] In some embodiments, the sending unit is further configured to send, to the second device, third indication information used to indicate a quasi-co-location relationship between the first reference signal and the second reference signal.

[0416] In some embodiments, the first reference signal and the second reference signal adopt different periodic configurations and / or different frequency domain densities.

[0417] In some embodiments, the first reference signal and the second reference signal satisfy one or more of the following: the first reference signal and the second reference signal are periodically sent reference signals, a period of the first reference signal is greater than a period of the second reference signal; a frequency domain density of the first reference signal is less than a frequency domain density of the second reference signal; the first reference signal is a periodically sent reference signal, and the second reference signal is a non-periodic or semi-persistent reference signal; the first reference signal is a non-periodic reference signal, and the second reference signal is a periodic or semi-persistent reference signal.

[0418] In some embodiments, the first channel information includes one or more of the following information: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of N, a LOS / NLOS indication, a multipath parameter of the channel; and / or the second channel information includes one or more of the following information: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of M.

[0419] In some embodiments, the third channel information comprises one or more of the following: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of N.

[0420] In some embodiments, the CSI comprises at least PMI information.

[0421] In some embodiments, the first frequency domain resource and the second frequency domain resource correspond to different bandwidths; or the first frequency domain resource and the second frequency domain resource contain different numbers of subbands; or the first frequency domain resource and the second frequency domain resource contain different numbers of subcarriers; or the first frequency domain resource and the second frequency domain resource contain different numbers of physical resource blocks (PRBs).

[0422] FIG. 13 is a schematic block diagram of a communication device according to an embodiment of the present application. The communication device 1300 shown in FIG. 13 can be any of the second devices described above. The second device can include a obtaining unit 1310.

[0423] The obtaining unit 1310 is configured to obtain, based on a first reference signal transmitted by a first device, first channel information corresponding to N antenna ports and / or a first frequency domain resource at a time t1, where the first reference signal corresponds to the N antenna ports and / or occupies the first frequency domain resource; and obtain, based on a second reference signal transmitted by the first device, second channel information corresponding to M antenna ports and / or a second frequency domain resource at a time t2, where the second reference signal corresponds to the M antenna ports and / or occupies the second frequency domain resource, M < N, the second frequency domain resource is part of the first frequency domain resource, and the time t2 is after the time t1. The determining unit 1320 is configured to determine, according to the first channel information and the second channel information, CSI corresponding to the N antenna ports and / or the first frequency domain resource at the time t2 or at a time t3, where the time t3 is after the time t2.

[0424] In some embodiments, the second device further includes an output unit configured to input the first channel information and the second channel information as inputs of a first AI model, and output, by the first AI model, the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the time t2.

[0425] In some embodiments, the determining unit is further configured to determine a second AI model according to the first channel information. The communication device further includes an output unit configured to input the second channel information as an input of the second AI model, and output, by the second AI model, the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the time t2.

[0426] In some embodiments, the communication device further comprises an output unit configured to input the first channel information and the second channel information as inputs of a third AI model, and output third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the t2 moment by the third AI model; and the determination unit is further configured to determine the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t2 moment according to the third channel information.

[0427] In some embodiments, the determination unit is further configured to determine a fourth AI model according to the first channel information; and the communication device further comprises an output unit configured to input the second channel information as an input of the fourth AI model, and output third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the t2 moment by the fourth AI model; and the determination unit is further configured to determine the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t2 moment according to the third channel information.

[0428] In some embodiments, the communication device further comprises an output unit configured to input the first channel information and second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different moments as inputs of a fifth AI model, and output the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment by the fifth AI model; and the multiple different moments of the second channel information corresponding to the M antenna ports and / or the second frequency domain resource include the t2 moment of the second channel information corresponding to the M antenna ports and / or the second frequency domain resource.

[0429] In some embodiments, the determination unit is further configured to determine a sixth AI model according to the first channel information; and the communication device further comprises an output unit configured to input second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different moments as inputs of the sixth AI model, and output the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment by the sixth AI model; and the multiple different moments of the second channel information corresponding to the M antenna ports and / or the second frequency domain resource include the t2 moment of the second channel information corresponding to the M antenna ports and / or the second frequency domain resource.

[0430] In some embodiments, the communication device further comprises an output unit configured to input the first channel information and second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different time instants into a seventh AI model, and output third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the t3 time instant through the seventh AI model; and the determination unit is further configured to determine the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t3 time instant according to the third channel information; wherein the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different time instants comprises the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at the t2 time instant.

[0431] In some embodiments, the determination unit is further configured to determine an eighth AI model according to the first channel information; and the communication device further comprises an output unit configured to input the first channel information and second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different time instants into the eighth AI model, and output third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the t3 time instant through the eighth AI model; and the determination unit is further configured to determine the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t3 time instant according to the third channel information; wherein the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different time instants comprises the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at the t2 time instant.

[0432] In some embodiments, the communication device further comprises a receiving unit configured to receive first indication information sent by the first device, the first indication information being used to indicate a mapping relationship between the M antenna ports and the N antenna ports; and the determination unit is further configured to determine the CSI corresponding to the N antenna ports at the t2 time instant or the t3 time instant according to the mapping relationship.

[0433] In some embodiments, the mapping relationship comprises any one of the following: the M antenna ports are obtained by beamforming the N antenna ports; the M antenna ports are obtained by performing horizontal dimension sampling on the N antenna ports; the M antenna ports are obtained by performing vertical dimension sampling on the N antenna ports; the M antenna ports are the first M antenna ports of the N antenna ports; the M antenna ports are the last M antenna ports of the N antenna ports; or the M antenna ports are the {n*k+1, n*k+2,.., n*k+M / 2}th ports in the N antenna ports, where k = 0, 1, and n = N / 2.

[0434] In some embodiments, the determining unit is further configured to determine, by the second device, an AI model used for calculating the CSI according to the mapping relationship; or the output unit is further configured to output the CSI or third channel information used for calculating the CSI by taking the mapping relationship as an input parameter of the AI model used for calculating the CSI.

[0435] In some embodiments, the communication device further includes a sending unit configured to send second indication information to the first device, the second indication information being used to instruct the first device to send a third reference signal, the third reference signal corresponding to N antenna ports and / or occupying the first frequency domain resource.

[0436] In some embodiments, the communication device further includes a receiving unit configured to receive the third reference signal sent by the first device; and an updating unit configured to update the first channel information based on the third reference signal.

[0437] In some embodiments, the first reference signal and the second reference signal are quasi co-located.

[0438] In some embodiments, the first reference signal and the second reference signal have different periodicity configurations and / or different frequency domain densities.

[0439] In some embodiments, the first reference signal and the second reference signal satisfy one or more of the following: the first reference signal and the second reference signal are periodically sent reference signals, a period of the first reference signal is greater than a period of the second reference signal; a frequency domain density of the first reference signal is less than a frequency domain density of the second reference signal; the first reference signal is a periodically sent reference signal, and the second reference signal is a non-periodic or semi-persistent reference signal; the first reference signal is a non-periodic reference signal, and the second reference signal is a periodic or semi-persistent reference signal.

[0440] In some embodiments, the first channel information includes one or more of the following information: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of N, a LOS / NLOS indication, a multipath parameter of the channel; and / or the second channel information includes one or more of the following information: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of M.

[0441] In some embodiments, the third channel information includes one or more of the following information: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of N.

[0442] In some embodiments, the CSI at least includes PMI information.

[0443] In some embodiments, the first frequency domain resource and the second frequency domain resource correspond to different bandwidths; or the first frequency domain resource and the second frequency domain resource contain different numbers of subbands; or the first frequency domain resource and the second frequency domain resource contain different numbers of subcarriers; or the first frequency domain resource and the second frequency domain resource contain different numbers of physical resource blocks (PRBs).

[0444] In some embodiments, the communication device further includes a sending unit configured to send the CSI to the first device, or a precoding unit configured to perform precoding on data according to the CSI, and a sending unit configured to send the precoded data to the first device.

[0445] FIG. 14 is a schematic structural diagram of a communication device according to an embodiment of the present application. The dashed line in FIG. 14 indicates that the unit or module is optional. The device 1400 can be used to implement the method described in the above method embodiments. The device 1400 can be a chip, a terminal device, or a network device.

[0446] The device 1400 can include one or more processors 1410. The processor 1410 can support the device 1400 to implement the method described in the above method embodiments. The processor 1410 can be a general purpose processor or a dedicated processor. For example, the processor can be a central processing unit (CPU). Alternatively, the processor can also be other general purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic components, discrete hardware components, etc. The general purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0447] The device 1400 can further include one or more memories 1420. The memory 1420 stores a program, which can be executed by the processor 1410, so that the processor 1410 performs the method described in the above method embodiments. The memory 1420 can be independent of the processor 1410 or integrated in the processor 1410.

[0448] The device 1400 can further include a transceiver 1430. The processor 1410 can communicate with other devices or chips through the transceiver 1430. For example, the processor 1410 can perform data transmission and reception with other devices or chips through the transceiver 1430.

[0449] The embodiment of the present application further provides a computer readable storage medium for storing a program. The computer readable storage medium can be applied to the terminal or the network device provided by the embodiment of the present application, and the program causes the computer to execute the method performed by the terminal or the network device in the various embodiments of the present application.

[0450] The embodiment of the present application further provides a computer program product. The computer program product includes a program. The computer program product can be applied to the terminal or the network device provided by the embodiment of the present application, and the program causes the computer to execute the method performed by the terminal or the network device in the various embodiments of the present application.

[0451] The embodiment of the present application further provides a computer program. The computer program can be applied to the terminal or the network device provided by the embodiment of the present application, and the computer program causes the computer to execute the method performed by the terminal or the network device in the various embodiments of the present application.

[0452] It should be understood that the terms "system" and "network" can be used interchangeably in the present application. In addition, the terms used in the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application. The terms "first", "second", "third", and "fourth" and the like in the description and claims of the present application and the drawings are used to distinguish different objects, and are not used to describe a particular order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion.

[0453] In the embodiments of the present application, the "indication" mentioned can be direct indication, or indirect indication, or can be an indication with an associated relationship. For example, A indicates B, which can mean that B can be obtained by A; or A indirectly indicates B, for example, A indicates C, and B can be obtained by C; or A and B have an associated relationship.

[0454] In the embodiments of the present application, "B corresponding to A" means that B is associated with A, and B can be determined according to A. However, it should also be understood that determining B according to A does not mean that B is determined only according to A, but B can also be determined according to A and / or other information.

[0455] In the embodiments of the present application, the term "corresponding" can mean that there is a direct or indirect corresponding relationship between the two, or can mean that there is an associated relationship between the two, or can mean an indication and being indicated, configuration and being configured, and the like.

[0456] In the embodiments of the present application, the "predefined" or "preconfigured" can be implemented by pre-storing corresponding codes, tables or other manners that can be used to indicate relevant information in devices (for example, including terminal devices and network devices), and the specific implementation manners are not limited in the present application. For example, the predefinition can refer to the definition in a protocol.

[0457] In the embodiments of the present application, the "protocol" can refer to a standard protocol in the communication field, for example, can include the LTE protocol, the NR protocol and the related protocol applied to the future communication system, and the present application is not limited to this.

[0458] In the embodiments of the present application, the term "and / or" is only used to describe the association relationship of the associated objects, and can represent three relationships, for example, A and / or B can represent the following three cases: A exists alone, A and B exist together, and B exists alone. In addition, the character " / " in the present application generally represents an "or" relationship between the front and rear associated objects.

[0459] In various embodiments of the present application, the size of the serial number of the above processes does not mean the order of execution, and the execution order of the processes should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0460] In several embodiments provided by the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only schematic. The division of the units is only a logical function division. There can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed coupling or direct coupling or communication connection between each other can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or in other forms.

[0461] The units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0462] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically, or two or more units can be integrated into one unit.

[0463] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) mode. The computer readable storage medium can be any available medium that can be read by a computer or a data storage device such as a server, data center and the like integrated with one or more available media sets. The available media can be magnetic media (for example, floppy disk, hard disk, magnetic tape), optical media (for example, digital video disc (DVD)) or semiconductor media (for example, solid state disk (SSD)) and the like.

[0464] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method of wireless communication, the method comprising: Comprise: The second device obtains first channel information corresponding to N antenna ports and / or first frequency domain resources at t1 time based on the first reference signal sent by the first device, wherein the first reference signal corresponds to N antenna ports and / or occupies the first frequency domain resources; The second device obtains second channel information corresponding to M antenna ports and / or second frequency domain resources at t2 time based on the second reference signal sent by the first device, wherein the second reference signal corresponds to the M antenna ports and / or occupies the second frequency domain resources, M < N, the second frequency domain resources are part of the first frequency domain resources, and t2 time is after t1 time; The second device determines channel state information (CSI) corresponding to N antenna ports and / or first frequency domain resources at t2 time or t3 time according to the first channel information and the second channel information, wherein t3 time is after t2 time.

2. The method of claim 1, wherein, The second device determines channel state information (CSI) corresponding to N antenna ports and / or first frequency domain resources at t2 time according to the first channel information and the second channel information, comprising: The second device inputs the first channel information and the second channel information as inputs of a first artificial intelligence (AI) model, and outputs the CSI corresponding to N antenna ports and / or first frequency domain resources at t2 time through the first AI model.

3. The method of claim 1, wherein, The second device determines channel state information (CSI) corresponding to N antenna ports and / or first frequency domain resources at t2 time according to the first channel information and the second channel information, comprising: The second device determines a second AI model according to the first channel information; The second device inputs the second channel information as an input of the second AI model, and outputs the CSI corresponding to N antenna ports and / or first frequency domain resources at t2 time through the second AI model.

4. The method of claim 1, wherein, The second device determines channel state information (CSI) corresponding to N antenna ports and / or first frequency domain resources at t2 time according to the first channel information and the second channel information, comprising: The second device inputs the first channel information and the second channel information as inputs of a third AI model, to output third channel information corresponding to N antenna ports and / or first frequency domain resources at t2 time through the third AI model; The second device determines the CSI corresponding to N antenna ports and / or first frequency domain resources at t2 time according to the third channel information.

5. The method of claim 1, wherein, The second device determines channel state information (CSI) corresponding to N antenna ports and / or first frequency domain resources at t2 time according to the first channel information and the second channel information, comprising: The second device determines a fourth AI model according to the first channel information; The second device inputs the second channel information as an input of the fourth AI model, to output third channel information corresponding to N antenna ports and / or first frequency domain resources at t2 time through the fourth AI model; The second device determines the CSI corresponding to N antenna ports and / or first frequency domain resources at t2 time according to the third channel information.

6. The method of claim 1, wherein, The second device determines channel state information (CSI) corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment according to the first channel information and the second channel information, including: The second device inputs the first channel information and second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different moments as input of the fifth AI model, so as to output the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment through the fifth AI model. The second device determines channel state information (CSI) corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment according to the first channel information and the second channel information, including:

7. The method of claim 1, wherein, The second device determines a sixth AI model according to the first channel information; The second device inputs the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different moments as input of the sixth AI model, so as to output the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment through the sixth AI model. The second device determines channel state information (CSI) corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment according to the first channel information and the second channel information, including: The second device inputs the first channel information and second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different moments as input of the seventh AI model, so as to output the third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment through the seventh AI model; 8. The method of claim 1, wherein, The second device determines the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment according to the third channel information; The second device determines channel state information (CSI) corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment according to the first channel information and the second channel information, including: The second device determines an eighth AI model according to the first channel information; The second device inputs the first channel information and second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different moments as input of the eighth AI model, so as to output the third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment through the eighth AI model; 9. The method of claim 1, wherein, The second device determines the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment according to the third channel information; ​ ​ ​ The second channel information corresponding to the M antenna ports and / or the second frequency domain resource at different time instants includes second channel information corresponding to the M antenna ports and / or the second frequency domain resource at the t2 time instant.

10. The method according to any one of claims 1-9, characterized in that, The method further includes: The second device receives first indication information sent by the first device, and the first indication information is used to indicate a mapping relationship between the M antenna ports and the N antenna ports. The second device determines the CSI corresponding to the N antenna ports at the t2 time instant or the t3 time instant according to the mapping relationship.

11. The method of claim 10, wherein, The mapping relationship includes any one of the following: The M antenna ports are obtained by beamforming from the N antenna ports; The M antenna ports are obtained by horizontal dimension sampling from the N antenna ports; The M antenna ports are obtained by vertical dimension sampling from the N antenna ports; The M antenna ports are the first M antenna ports of the N antenna ports; The M antenna ports are the last M antenna ports of the N antenna ports; Or The M antenna ports are the {n*k+1, n*k+2,..., n*k+M / 2}th ports in the N antenna ports, where k=0, 1, and n=N / 2.

12. The method according to claim 10 or 11, characterized in that, The second device determines the CSI corresponding to the N antenna ports at the t2 time instant or the t3 time instant according to the mapping relationship, including: The second device determines an AI model used when calculating the CSI according to the mapping relationship; or The second device takes the mapping relationship as an input parameter of the AI model used when calculating the CSI to output the CSI or output third channel information used for calculating the CSI.

13. The method according to any one of claims 1-12, characterized in that, The method further includes: The second device sends second indication information to the first device, and the second indication information is used to instruct the first device to send a third reference signal, and the third reference signal corresponds to the N antenna ports and / or occupies the first frequency domain resource.

14. The method of claim 13, wherein, The method further includes: The second device receives the third reference signal sent by the first device; The second device updates the first channel information based on the third reference signal.

15. The method of any one of claims 1-14, wherein, The first reference signal and the second reference signal are quasi-co-located.

16. The method of any one of claims 1-15, wherein, The first reference signal and the second reference signal adopt different periodic configurations and / or different frequency domain densities.

17. The method of claim 16, wherein, The first reference signal and the second reference signal satisfy one or more of the following: The first reference signal and the second reference signal are periodically transmitted reference signals, and the period of the first reference signal is greater than the period of the second reference signal; The frequency domain density of the first reference signal is less than the frequency domain density of the second reference signal; The first reference signal is a periodically transmitted reference signal, and the second reference signal is a non-periodic or semi-persistent reference signal; The first reference signal is a non-periodic reference signal, and the second reference signal is a periodic or semi-persistent reference signal.

18. The method of any one of claims 1-17, wherein, The first channel information comprises one or more of the following: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of N, a line-of-sight (LOS) / non-line-of-sight (NLOS) indication, a multipath parameter of the channel; and / or The second channel information comprises one or more of the following: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of M.

19. The method of any one of claims 4, 5, 8, 9, 12, 14, wherein, The third channel information comprises one or more of the following: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of N.

20. The method of any one of claims 1-19, wherein, The CSI at least comprises precoding matrix indication (PMI) information.

21. The method of any one of claims 1-20, wherein, The first frequency domain resource and the second frequency domain resource correspond to different bandwidths; or The first frequency domain resource and the second frequency domain resource comprise different numbers of subbands; or The first frequency domain resource and the second frequency domain resource comprise different numbers of subcarriers; or The first frequency domain resource and the second frequency domain resource comprise different numbers of physical resource blocks (PRBs).

22. The method of any one of claims 1-21, wherein, The method further comprises: The second device sends the CSI to the first device, or The second device performs precoding on data according to the CSI and sends the precoded data to the first device.

23. A method of wireless communication, the method comprising: Comprise: The first device sends a first reference signal at time t1, the first reference signal corresponding to N antenna ports and / or occupying a first frequency domain resource; The first device sends a second reference signal at time t2, the second reference signal corresponding to M antenna ports and / or occupying a second frequency domain resource, where M < N, and time t2 is after time t1; The first device receives CSI sent by the second device, or the first device receives precoded data sent by the second device, the precoded data being data precoded based on the CSI, the CSI corresponding to N antenna ports and / or the first frequency domain resource at time t2 or time t3, the CSI being measured based on the first reference signal and the second reference signal, where time t3 is after time t2.

24. The method of claim 23, wherein, The CSI is measured based on the first reference signal and the second reference signal, comprising: The CSI is determined based on first channel information corresponding to N antenna ports and / or the first frequency domain resource and second channel information corresponding to M antenna ports and / or the second frequency domain resource, where the first channel information is measured based on the first reference signal, and the second channel information is measured based on the second reference signal.

25. The method of claim 23 or 24, wherein, The M antenna ports used to send the second reference signal have a mapping relationship with the N antenna ports used to send the first reference signal as follows: The M antenna ports are obtained by beamforming the N antenna ports; or The M antenna ports are obtained by horizontally sampling the N antenna ports; or The M antenna ports are obtained by vertically sampling the N antenna ports; or The M antenna ports are the first M antenna ports of the N antenna ports; or The M antenna ports are the last M antenna ports of the N antenna ports; or ​ ​ The M antenna ports are the {n*k+1, n*k+2,..,n*k+M / 2}th ports in the N antenna ports, where k=0, 1, n=N / 2.

26. The method of any one of claims 23-25, wherein, The method further comprises: The first device sends first indication information to the second device, the first indication information being used to indicate a mapping relationship between the M antenna ports and the N antenna ports, and the first indication information being used for the second device to determine the CSI corresponding to the N antenna ports at the t2 moment or the t3 moment according to the mapping relationship.

27. The method of claim 26, wherein, The mapping relationship is used for the second device to determine an AI model used when calculating the CSI; or The mapping relationship is used as an input parameter of the AI model used when calculating the CSI to output the CSI or output third channel information used for calculating the CSI.

28. The method of any one of claims 23-27, wherein, The method further comprises: The first device receives second indication information sent by the second device, the second indication information being used to instruct the first device to send a third reference signal, and the third reference signal corresponding to the N antenna ports and / or occupying a first frequency domain resource; The first device sends the third reference signal.

29. The method of any one of claims 23-28, wherein, The method further comprises: The first device sends third indication information to the second device, the third indication information being used to indicate a quasi-co-location relationship between the first reference signal and the second reference signal.

30. The method of any one of claims 23-29, wherein, The first reference signal and the second reference signal adopt different periodic configurations and / or different frequency domain densities.

31. The method of claim 30, wherein, The first reference signal and the second reference signal satisfy one or more of the following conditions: The first reference signal and the second reference signal are periodically transmitted reference signals, and the period of the first reference signal is greater than the period of the second reference signal; The frequency domain density of the first reference signal is less than the frequency domain density of the second reference signal; The first reference signal is a periodically transmitted reference signal, and the second reference signal is a non-periodic or semi-persistent reference signal; The first reference signal is a non-periodic reference signal, and the second reference signal is a periodic or semi-persistent reference signal.

32. The method of any one of claims 23-31, wherein, The first channel information includes one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, value of N, line of sight (LOS) / non-line of sight (NLOS) indication, and multipath parameter of the channel. And / or The second channel information includes one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, and value of M.

33. The method of claim 27, wherein, The third channel information includes one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, and value of N.

34. The method of any one of claims 23-33, wherein, The CSI at least includes precoding matrix indication (PMI) information.

35. The method of any one of claims 23-34, wherein, The first frequency domain resource and the second frequency domain resource correspond to different bandwidths; or The first frequency domain resource and the second frequency domain resource contain different numbers of subbands; or The first frequency domain resource and the second frequency domain resource contain different numbers of subcarriers; or The first frequency domain resource and the second frequency domain resource contain different numbers of physical resource blocks (PRBs).

36. A communications device, characterized by The communication device is a second device, which comprises: obtain, based on a first reference signal sent by a first device, first channel information corresponding to N antenna ports and / or a first frequency domain resource at a time t1, wherein the first reference signal corresponds to the N antenna ports and / or occupies the first frequency domain resource; obtain, based on a second reference signal sent by the first device, second channel information corresponding to M antenna ports and / or a second frequency domain resource at a time t2, wherein the second reference signal corresponds to the M antenna ports and / or occupies the second frequency domain resource, M < N, the second frequency domain resource is part of the first frequency domain resource, and the time t2 is after the time t1; determine, according to the first channel information and the second channel information, channel state information (CSI) corresponding to the N antenna ports and / or the first frequency domain resource at a time t3, wherein the time t3 is after the time t2.

37. The communication device of claim 36, wherein, The communication device further comprises: output, as input of a first artificial intelligence (AI) model, the first channel information and the second channel information, and output, by the first AI model, the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the time t2.

38. The communication device of claim 37, wherein the determining unit is further configured to determine a second AI model according to the first channel information; the communication device further comprises an output unit configured to output, as input of the second AI model, the second channel information, and output, by the second AI model, the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the time t2.

39. The communication device of claim 38, wherein the determining unit determines, according to the first channel information and the second channel information, the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the time t2, and the second device further comprises: the communication device further comprises an output unit configured to output, as input of a third AI model, the first channel information and the second channel information, and output, by the third AI model, third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the time t2; the determining unit is further configured to determine, according to the third channel information, the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the time t2.

40. The communication device of claim 39, wherein the determining unit is further configured to determine a fourth AI model according to the first channel information; the communication device further comprises an output unit configured to output, as input of the fourth AI model, the second channel information, and output, by the fourth AI model, third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the time t2; the determining unit is further configured to determine, according to the third channel information, the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the time t2.

41. The communication device of claim 40, wherein, the communication device further comprises: an output unit, configured to input the first channel information and second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different time moments as input of a fifth AI model, and output, through the fifth AI model, the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment; wherein the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different time moments includes the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at the t2 moment.

42. The communication device of claim 41, wherein the determination unit is further configured to determine a sixth AI model according to the first channel information; the communication device further comprises an output unit, configured to input the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different time moments as input of the sixth AI model, and output, through the sixth AI model, the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment. wherein the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different time moments includes the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at the t2 moment.

43. The communication device of claim 42, wherein the communication device further comprises an output unit, configured to input the first channel information and second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different time moments as input of a seventh AI model, and output, through the seventh AI model, third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment; the determination unit is further configured to determine the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment according to the third channel information; wherein the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different time moments includes the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at the t2 moment.

44. The communication device of claim 43, wherein the determination unit is further configured to determine an eighth AI model according to the first channel information; the communication device further comprises an output unit, configured to input the first channel information and second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different time moments as input of the eighth AI model, and output, through the eighth AI model, third channel information corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment; the determination unit is further configured to determine the CSI corresponding to the N antenna ports and / or the first frequency domain resource at the t3 moment according to the third channel information; wherein the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at multiple different time moments includes the second channel information corresponding to the M antenna ports and / or the second frequency domain resource at the t2 moment.

45. The communication device of any one of claims 36-44, wherein The communication device further includes a receiving unit configured to receive first indication information sent by the first device, the first indication information being used to indicate a mapping relationship between the M antenna ports and the N antenna ports. The determining unit is further configured to determine, according to the mapping relationship, CSI corresponding to the N antenna ports at the t2 moment or the t3 moment.

46. The communication device of claim 45, wherein, The mapping relationship includes any one of the following: The M antenna ports are obtained by beamforming from the N antenna ports; The M antenna ports are obtained by horizontal dimension sampling from the N antenna ports; The M antenna ports are obtained by vertical dimension sampling from the N antenna ports; The M antenna ports are the first M antenna ports of the N antenna ports; The M antenna ports are the last M antenna ports of the N antenna ports; Or The M antenna ports are the {n*k+1, n*k+2,.., n*k+M / 2}th ports in the N antenna ports, where k=0, 1, and n=N / 2.

47. The communication device of claim 45 or 46, wherein: The determining unit is further configured to determine, according to the mapping relationship, an AI model used when calculating the CSI; or The communication device further includes an output unit configured to output the CSI or output third channel information used to calculate the CSI, by taking the mapping relationship as an input parameter of the AI model used when calculating the CSI.

48. The communication device of any of claims 36-47, wherein, The communication device further includes: a sending unit configured to send, to the first device, second indication information used to instruct the first device to send a third reference signal, the third reference signal corresponding to the N antenna ports and / or occupying the first frequency domain resource.

49. The communication device of claim 48, wherein, The communication device further includes: a receiving unit configured to receive the third reference signal sent by the first device; an updating unit configured to update the first channel information based on the third reference signal.

50. The communication device of any of claims 36-49, wherein, The first reference signal and the second reference signal are quasi-co-located.

51. The communication device of any of claims 36-50, wherein, The first reference signal and the second reference signal have different periodicity configurations and / or different frequency domain densities.

52. The communication device of claim 51, wherein, The first reference signal and the second reference signal satisfy one or more of the following: The first reference signal and the second reference signal are periodically sent reference signals, and the period of the first reference signal is greater than the period of the second reference signal; The frequency domain density of the first reference signal is less than the frequency domain density of the second reference signal; The first reference signal is a periodically sent reference signal, and the second reference signal is a non-periodic or semi-persistent reference signal; The first reference signal is a non-periodic reference signal, and the second reference signal is a periodic or semi-persistent reference signal.

53. The communication device of any of claims 36-52, wherein, The first channel information includes one or more of the following information: channel matrix, channel covariance matrix, channel eigenvector, value of N, line of sight (LOS) / non-line of sight (NLOS) indication, and multipath parameter of the channel. And / or The second channel information comprises one or more of the following: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of M.

54. The communication device of any of claims 39, 40, 43, 44, 47, 49, wherein, The third channel information comprises one or more of the following: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of N.

55. The communication device of any of claims 36-54, wherein, The CSI comprises at least precoding matrix indication (PMI) information.

56. The communication device of any of claims 36-55, wherein, The first frequency domain resource and the second frequency domain resource correspond to different bandwidths; or The first frequency domain resource and the second frequency domain resource comprise different numbers of subbands; or The first frequency domain resource and the second frequency domain resource comprise different numbers of subcarriers; or The first frequency domain resource and the second frequency domain resource comprise different numbers of physical resource blocks (PRBs).

57. The communication device of any of claims 36-56, wherein, The communication device further comprises: a sending unit configured to send the CSI to the first device, or a precoding unit configured to perform precoding on data according to the CSI, and a sending unit configured to send the precoded data to the first device.

58. A communications device, characterized by The communication device is a first device, comprising: a sending unit configured to send a first reference signal at time t1, the first reference signal corresponding to N antenna ports and / or occupying a first frequency domain resource; the sending unit is further configured to send a second reference signal at time t2, the second reference signal corresponding to M antenna ports and / or occupying a second frequency domain resource, wherein M < N, and time t2 is after time t1; a receiving unit configured to receive CSI sent by a second device, or a receiving unit configured to receive precoded data sent by the second device, the precoded data being data precoded based on the CSI, the CSI corresponding to N antenna ports and / or the first frequency domain resource at time t2 or time t3, the CSI being measured based on the first reference signal and the second reference signal, wherein time t3 is after time t2.

59. The communication device of claim 58, wherein, The CSI is measured based on the first reference signal and the second reference signal, comprising: The CSI is determined based on first channel information corresponding to N antenna ports and / or the first frequency domain resource and second channel information corresponding to M antenna ports and / or the second frequency domain resource, wherein the first channel information is measured based on the first reference signal, and the second channel information is measured based on the second reference signal.

60. The communication device of claim 58 or 59, wherein, The M antenna ports used to send the second reference signal have the following mapping relationship with the N antenna ports used to send the first reference signal: The M antenna ports are obtained by beamforming the N antenna ports; or The M antenna ports are obtained by horizontally sampling the N antenna ports; or The M antenna ports are obtained by vertically sampling the N antenna ports; or The M antenna ports are the first M antenna ports of the N antenna ports; or The M antenna ports are the last M antenna ports of the N antenna ports; or The M antenna ports are the {n*k+1, n*k+2,.., n*k+M / 2}th ports in the N antenna ports, where k = 0, 1, n = N / 2. ​ ​ 61. The communication device of any of claims 58-60, wherein, The sending unit is further configured to: send first indication information to the second device, the first indication information being used to indicate a mapping relationship between the M antenna ports and the N antenna ports, and the first indication information being used for the second device to determine the CSI corresponding to the N antenna ports at the t2 moment or the t3 moment according to the mapping relationship.

62. The communication device of claim 61, wherein, The mapping relationship is used for the second device to determine an AI model used when calculating the CSI; or The mapping relationship is used as an input parameter of the AI model used when calculating the CSI to output the CSI or output third channel information used for calculating the CSI.

63. The communication device of any one of claims 58-62, wherein The receiving unit is further configured to receive second indication information sent by the second device, the second indication information being used to indicate that the first device sends a third reference signal, and the third reference signal corresponding to N antenna ports and / or occupying the first frequency domain resource. The sending unit is further configured to send the third reference signal.

64. The communication device of any one of claims 58-63, wherein The sending unit is further configured to send third indication information to the second device, the third indication information being used to indicate a quasi-co-location relationship between the first reference signal and the second reference signal.

65. The communication device of any of claims 58-64, wherein, The first reference signal and the second reference signal adopt different periodic configurations and / or different frequency domain densities.

66. The communication device of claim 65, wherein, The first reference signal and the second reference signal satisfy one or more of the following: The first reference signal and the second reference signal are periodically transmitted reference signals, and a period of the first reference signal is greater than a period of the second reference signal. The frequency domain density of the first reference signal is less than the frequency domain density of the second reference signal. The first reference signal is a periodically transmitted reference signal, and the second reference signal is a non-periodic or semi-persistent reference signal. The first reference signal is a non-periodic reference signal, and the second reference signal is a periodic or semi-persistent reference signal.

67. The communication device of any of claims 58-66, wherein, The first channel information includes one or more of the following information: a channel matrix, a channel covariance matrix, a channel eigenvector, a value of N, a line of sight (LOS) / non-line of sight (NLOS) indication, and a channel multipath parameter. And / or The second channel information includes one or more of the following information: a channel matrix, a channel covariance matrix, a channel eigenvector, and a value of M.

68. The communications device of claim 62, wherein The third channel information includes one or more of the following information: a channel matrix, a channel covariance matrix, a channel eigenvector, and a value of N.

69. The communication device of any of claims 58-68, wherein, The CSI at least includes precoding matrix indication (PMI) information.

70. The communication device of any of claims 58-69, wherein, The first frequency domain resource and the second frequency domain resource correspond to different bandwidths; or The first frequency domain resource and the second frequency domain resource contain different numbers of subbands; or The first frequency domain resource and the second frequency domain resource contain different numbers of subcarriers; or The first frequency domain resource and the second frequency domain resource contain different numbers of physical resource blocks (PRBs).

71. A communications device, comprising: The communication device is a second device comprising a transceiver, a memory and a processor, the memory being configured to store a program, the processor being configured to invoke the program in the memory and control the transceiver to receive or send a signal, so that the terminal executes the method according to any one of claims 1-22.

72. A communications device, comprising: The communication device is a first device comprising a transceiver, a memory and a processor, the memory being configured to store a program, the processor being configured to invoke the program in the memory and control the transceiver to receive or send a signal, so that the network device executes the method according to any one of claims 23-35.

73. An apparatus comprising: A processor is configured to invoke a program from a memory, so that the apparatus executes the method according to any one of claims 1-35.

74. A chip, comprising: A processor is configured to invoke a program from a memory, so that the apparatus executes the method according to any one of claims 1-35.

75. A computer-readable storage medium, comprising: A computer program is stored on the computer-readable medium, and the computer program causes the computer to execute the method according to any one of claims 1-35.

76. A computer program product, characterized in that, A computer program is stored on the computer-readable medium, and the computer program causes the computer to execute the method according to any one of claims 1-35.

77. A computer program, characterized in that, The computer program causes the computer to execute the method according to any one of claims 1-35.

Images