Uplink precoding method and apparatus

Abstract

An uplink precoding method and apparatus are provided. The method includes: a terminal device receives first information from an access network device; the terminal device determines a first precoding matrix based on the first information and a first model, where the first precoding matrix is ​​an output of the first model, and the input of the first model is determined based on the first information; the terminal device precodes an uplink signal based on the first precoding matrix; and the terminal device transmits the precoded uplink signal to the access network device. In this method, the terminal device determines the first precoding matrix based on the first information and the first model. The precoding matrix is ​​no longer selected from an offline codebook, but is a floating-point matrix generated by the terminal device based on the first information. This can better fit the channel environment and improve the precoding effect.

Description

[Technical field]

[0001] FIELD OF THE DISCLOSURE Embodiments of the present disclosure relate to the field of communication technologies, and in particular to an uplink precoding method and apparatus. [Background technology]

[0002] [CROSS REFERENCE TO RELATED APPLICATIONS] This application claims priority to Chinese Patent Application No. 202111668221.2, entitled "UPLINK PRECODING METHOD AND APPARATUS," filed with the State Intellectual Property Office of China on December 31, 2021, the entire contents of which are incorporated herein by reference.

[0003] In wireless communication networks, such as mobile communication networks, the services supported by the network are becoming more and more diverse, and therefore the requirements that need to be met are becoming more and more diverse. For example, the network needs to be able to support very high rates, very low latency, and / or very large connections. This characteristic makes network planning, network configuration, and / or resource scheduling more and more complex. Furthermore, as the network has more and more powerful capabilities to support new technologies, such as, for example, higher and higher spectrum, and higher order multiple-input multiple-output (MIMO) technology, beamforming, and / or beam management, energy conservation in the network has become a hot research topic. These new requirements, new scenarios, and new features bring unprecedented challenges to network planning, operation and maintenance, and efficient operation. To address this challenge, artificial intelligence techniques may be introduced into wireless communication networks to implement network intelligence. Based on this, how to effectively implement artificial intelligence in networks has become a problem worth studying. Summary of the Invention

[0004] The present disclosure provides an uplink precoding method and apparatus for performing precoding on an uplink signal.

[0005] According to a first aspect, an uplink precoding method is provided. The method may be executed by a terminal device or may be executed by a component (e.g., a processor, a chip, or another component) configured in the terminal device, or may be executed by a software module, etc. The method includes receiving first information from an access network device, precoding an uplink signal based on a first precoding matrix, where the first precoding matrix is ​​determined based on an output of a first model, and an input of the first model is determined based on the first information, and transmitting the precoded uplink signal to the access network device.

[0006] According to the above method, the terminal device determines a first precoding matrix based on the first information and the first model. The precoding matrix is ​​no longer selected from an offline codebook, but may be a floating-point matrix generated by the terminal device based on the first information. Because the floating-point matrix is ​​a complex matrix, the generated precoding matrix is ​​better adapted to the channel environment, thereby improving the precoding effect.

[0007] In one design, the first information indicates a compressed first precoding matrix, and the output of the first model includes the first precoding matrix. Alternatively, the first information indicates compressed channel state information (CSI), the output of the first model includes uplink channel information, and the first precoding matrix is ​​determined based on the uplink channel information. Alternatively, the first information indicates compressed CSI, and the output of the first model includes the first precoding matrix.

[0008] In this disclosure, compressed CSI may alternatively be referred to as CSI.

[0009] In one design, the first precoding matrix may be at a wideband level, a subband level, or a resource element level.

[0010] According to the above design, if the first precoding matrix generated by the terminal device is at the bandwidth level, the uplink signals of the entire bandwidth are precoded with the same precoding matrix. Alternatively, if the first precoding matrix is ​​at the subband level, the uplink signals are precoded with precoding matrices corresponding to different subbands of the entire bandwidth. Similarly, if the first precoding matrix is ​​at the resource element RE level, the uplink signals of each RE are precoded with uplink signals corresponding to the uplink signals of each RE. Compared with the bandwidth-level precoding matrix, the subband-level precoding matrix or the RE-level precoding matrix can improve the codebook and channel adaptation, and can improve the precoding accuracy.

[0011] According to a second aspect, an uplink precoding method is provided. The method may be executed by an access network device or may be executed by a component (e.g., a processor, a chip, or another component) configured in the access network device, or may be executed by a software module, etc. The method includes determining first information based on uplink channel information and a second model, an input of the second model being determined based on the uplink channel information, and the first information being determined based on an output of the second model, and transmitting the first information to a terminal device, the first information indicating a compressed first precoding matrix or the first information indicating compressed channel state information CSI.

[0012] According to the above design, the second model is deployed in an access network device, the access network device may determine first information based on uplink channel information of the terminal device and the second model, and the terminal device may restore a precoding matrix based on the first information and the first model. The precoding matrix restored by the terminal device based on the first information is a floating-point matrix rather than being selected from an offline matrix, which better fits the current channel environment of the terminal device, thereby improving precoding accuracy.

[0013] According to a third aspect, an uplink precoding method is provided. The method may be executed by an access network device or by a component (e.g., a processor, a chip, or another component) configured in the access network device, or may be executed by a software module, etc. The method includes receiving a first reference signal from a terminal device, determining a first codebook based on the received first reference signal, where the first codebook is used to precode an uplink signal, and transmitting the first codebook to the terminal device.

[0014] According to the above method, the access network device may configure a corresponding personalized codebook for the terminal device, i.e., the above-mentioned first codebook, according to the reference signal transmitted by the terminal device.Compared with the fixed codebook used by the terminal device, the personalized codebook can better fit the channel characteristics of each terminal device, thereby improving the precoding accuracy of the terminal device.

[0015] In one design, determining the first codebook based on the received first reference signal includes separately measuring the first reference signal on Z resources to obtain Z pieces of uplink channel information, where Z is an integer greater than or equal to one, and determining the first codebook based on the Z pieces of uplink channel information.

[0016] In one design, determining the first codebook based on the Z pieces of uplink channel information includes determining the first codebook based on the Z pieces of uplink channel information and a codebook generation model.

[0017] In one design, determining the first codebook based on the Z uplink channel information includes determining Z precoding matrices based on the Z uplink channel information and determining the first codebook based on the Z precoding matrices.

[0018] In one design, determining the first codebook based on the Z precoding matrices includes determining M precoding matrices based on the Z precoding matrices and a precoding matrix generation model, where M is less than or equal to Z, and the first codebook includes the M precoding matrices.

[0019] In one design, determining the first codebook based on the Z precoding matrices includes determining M precoding matrices based on the Z precoding matrices and a precoding matrix generation model, where M is less than or equal to N, and separately quantizing the M precoding matrices to obtain M quantized precoding matrices, where the first codebook includes the M quantized precoding matrices.

[0020] According to the above design, when configuring a personalized codebook for a terminal device, the access network device determines Z precoding matrices based on the collected Z uplink channel information of the terminal device. The Z coding matrices are converged to M precoding matrices, where the value of M is less than or equal to the value of Z. The personalized codebook configured for the terminal device includes M precoding matrices. This design requires the access network device to configure or indicate the personalized codebook to the terminal device. Thus, when the above design is used, overhead for indicating the personalized codebook by the terminal device may be reduced.

[0021] In one design, the method further includes sending indication information to the terminal device, where the indication information indicates X precoding matrices corresponding to the X subbands, the X precoding matrices belonging to a first codebook, each of the X precoding matrices being used to precode an uplink signal on a corresponding subband, where X is a positive integer.

[0022] According to the above design, after configuring the personalized codebook for the terminal device, the access network device may further send indication information to the terminal device, where the indication information indicates a precoding matrix corresponding to each subband in the uplink bandwidth of the terminal device. Compared with a method in which the same precoding matrix is ​​used in the entire uplink bandwidth, in this method, the precoding matrix can be adapted to the channel characteristics corresponding to each subband, thereby improving the precoding accuracy.

[0023] In one design, transmitting the display information to the terminal device includes transmitting the display information to the terminal device according to a display rule of the subband precoding matrix.

[0024] In one design, the method further includes transmitting the representation rule of the subband precoding matrix to a terminal device.

[0025] In one design, the subband precoding matrix display rule includes: an uplink channel bandwidth includes at least a first subband and a second subband, where the display rule of the precoding matrix corresponding to the first subband includes indicating an index of the precoding matrix corresponding to the first subband, and the display rule of the precoding matrix corresponding to the second subband includes indicating an offset direction and an offset value of an index of the precoding matrix of the second subband relative to an index of the precoding matrix of the first subband.

[0026] According to the above design, if the access network device indicates the index of the precoding matrix corresponding to each subband separately, the indication overhead is high. In this design, the access network device predefines an indication rule and informs the terminal device of the indication rule. Then, the access network device indicates the corresponding precoding matrix index for some subbands of the entire bandwidth. Another subband may serve as an adjacent band of these some subbands. The access network device indicates the offset direction and offset value of the precoding matrix index of this adjacent band relative to the precoding matrix index corresponding to the previous subband, etc., thereby reducing the overhead for indicating the precoding matrix.

[0027] In one design, the first codebook is a codebook for a first subband, and the first codebook is used to precode an uplink signal on the first subband.

[0028] According to the above design, the base station may configure a codebook applicable to the entire uplink bandwidth, i.e., a first codebook, for the terminal device. Then, the base station may need to indicate to the terminal device an index of a precoding matrix corresponding to each subband in the first codebook. Alternatively, the base station may directly configure a codebook applicable to each subband, e.g., a first codebook applicable to the first subband, for the terminal. Then, the base station indicates an index of a precoding matrix corresponding to each subband in the codebook corresponding to each subband, etc. This manner of configuring a codebook corresponding to each subband may improve the accuracy of the codebook configured for the terminal.

[0029] According to a fourth aspect, an uplink precoding method is provided. The method corresponds to the terminal device in the third aspect. For advantageous effects, please refer to the third aspect. The method may be executed by the terminal device or by a component (e.g., a processor, a chip, or another component) configured in the terminal device, or may be executed by a software module or the like. The method includes transmitting a first reference signal to an access network device, receiving a first codebook from the access network device, and precoding an uplink signal using the first codebook.

[0030] In one design, precoding the uplink signal with the first codebook includes receiving indication information from the access network device, the indication information indicating X precoding matrices corresponding to the X subbands, the X precoding matrices belonging to the first codebook, and precoding the uplink signal on the corresponding subband with each of the X precoding matrices.

[0031] In one design, the method further includes receiving a subband precoding matrix indication rule from the access network device.

[0032] In one design, the subband precoding matrix display rule includes: an uplink channel bandwidth includes at least a first subband and a second subband, where the display rule of the precoding matrix corresponding to the first subband includes indicating an index of the precoding matrix corresponding to the first subband, and the display rule of the precoding matrix corresponding to the second subband includes indicating an offset direction and an offset value of an index of the precoding matrix of the second subband relative to an index of the precoding matrix of the first subband.

[0033] In one design, the first codebook is a codebook for a first subband, and precoding the uplink signal with the first codebook includes precoding the uplink signal of the first subband with the first codebook.

[0034] According to a fifth aspect, an apparatus is provided. For advantageous effects, see the description of the first or fourth aspect. The apparatus may be a terminal device, an apparatus configured in a terminal device, or an apparatus capable of being used in combination with a terminal device. In one design, the apparatus may include units in one-to-one correspondence with the methods / operations / steps / actions described in the first or fourth aspect. These units may be implemented using hardware circuits, software, or a combination of hardware circuits and software.

[0035] For example, the apparatus may include a processing unit and a communication unit, and the processing unit and the communication unit may perform corresponding functions in any design example of the first aspect. Specifically, as follows.

[0036] The communication unit is configured to receive first information from the access network device. The processing unit is configured to precode the uplink signal based on a first precoding matrix, the first precoding matrix being determined based on an output of the first model, and an input of the first model being determined based on the first information. The communication unit is further configured to transmit the precoded uplink signal to the access network device.

[0037] For the specific execution processes of the processing unit and the communication unit, please refer to the description of the first embodiment.

[0038] For example, the apparatus includes a processor configured to perform the method described in the first aspect. The apparatus may further include a memory configured to store instructions and / or data. The memory is coupled to the processor. When the processor executes the program instructions stored in the memory, the method of the first aspect may be performed. The apparatus may further include a communication interface, which is used for the apparatus to communicate with another device. For example, the communication interface may be a transceiver, a circuit, a bus, a module, a pin, or another type of communication interface, and the other device may be an access network device, etc.

[0039] In a possible design, the apparatus is configured to carry out the method described in the first aspect. The apparatus comprises: a memory configured to store program instructions; a communication interface configured to receive first information from the access network device and to transmit a precoded uplink signal to the access network device; and a processor configured to precode the uplink signal based on a first precoding matrix, the first precoding matrix being determined based on an output of a first model, and an input of the first model being determined based on first information.

[0040] For the specific execution process of the communication interface and the processor, please refer to the description of the first embodiment, and the details will not be described again.

[0041] For example, the apparatus may include a processing unit and a communication unit, and the processing unit and the communication unit may perform corresponding functions in any design example of the fourth aspect. Specifically, as follows.

[0042] The communication unit is configured to transmit a first reference signal to the access network device and to receive a first codebook from the access network device. The processing unit is configured to precode an uplink signal using the first codebook.

[0043] For the specific execution process of the processing unit and the communication unit, please refer to the description of the fourth embodiment.

[0044] For example, the apparatus includes a processor configured to perform the method described in the fourth aspect. The apparatus may further include a memory configured to store instructions and / or data. The memory is coupled to the processor. When the processor executes the program instructions stored in the memory, the method of the fourth aspect may be performed. The apparatus may further include a communication interface, which is used for the apparatus to communicate with another device. For example, the communication interface may be a transceiver, a circuit, a bus, a module, a pin, or another type of communication interface, and the other device may be an access network device, etc.

[0045] In a possible design, the apparatus is configured to carry out the method described in the fourth aspect. a memory configured to store program instructions; a communication unit configured to transmit a first reference signal to an access network device and receive a first codebook from the access network device; and a processor configured to precode the uplink signal using the first codebook.

[0046] For the specific execution process of the communication interface and the processor, please refer to the description of the fourth embodiment, and the details will not be described again.

[0047] According to a sixth aspect, an apparatus is provided. For advantageous effects, see the description of the second or third aspect. The apparatus may be an access network device, an apparatus configured in an access network device, or an apparatus capable of being used in combination with an access network device. In one design, the apparatus may include units in one-to-one correspondence with the methods / operations / steps / actions described in the second or third aspect. These units may be implemented using hardware circuits, software, or a combination of hardware circuits and software.

[0048] For example, the apparatus may include a processing unit and a communication unit, and the processing unit and the communication unit may perform corresponding functions in any design example of the second aspect. Specifically, as follows.

[0049] The processing unit is configured to determine the first information based on the uplink channel information and a second model, an input of the second model being determined based on the uplink channel information, and the first information being determined based on an output of the second model.

[0050] The communication unit is configured to transmit first information to the terminal device, where the first information indicates a compressed first precoding matrix, or the first information indicates compressed channel state information (CSI).

[0051] For the specific execution process of the processing unit and the communication unit, please refer to the description of the second embodiment, and the details will not be described again in this specification.

[0052] For example, the apparatus includes a processor configured to perform the method described in the second aspect. The apparatus may further include a memory configured to store instructions and / or data. The memory is coupled to the processor. When the processor executes the program instructions stored in the memory, the method described in the second aspect may be performed. The apparatus may further include a communication interface, which is used for the apparatus to communicate with another device. For example, the communication interface may be a transceiver, a circuit, a bus, a module, a pin, or another type of communication interface, and the other device may be a terminal, etc. In a possible design, the apparatus may include: a memory configured to store program instructions; a processor configured to determine first information based on uplink channel information and a second model, where an input of the second model is determined based on the uplink channel information and the first information is determined based on an output of the second model; and and a communication interface configured to transmit first information to a terminal device, the first information indicating a compressed first precoding matrix or the first information indicating compressed channel state information (CSI).

[0053] For the specific execution process of the communication interface and the processor, please refer to the description of the second embodiment, and the details will not be described again.

[0054] For example, the apparatus may include a processing unit and a communication unit, and the processing unit and the communication unit may perform corresponding functions in any design example of the third aspect. Specifically, as follows.

[0055] The communication unit is configured to receive a first reference signal from a terminal device.

[0056] The processing unit is configured to determine a first codebook based on the received first reference signal, where the first codebook is used for precoding the uplink signal.

[0057] The communication unit is configured to transmit the first codebook to the terminal device.

[0058] For the specific execution process of the communication unit and the processing unit, please refer to the description of the third embodiment, and the details will not be described again.

[0059] For example, the apparatus includes a processor configured to perform the method described in the third aspect. The apparatus may further include a memory configured to store instructions and / or data. The memory is coupled to the processor. When the processor executes the program instructions stored in the memory, the method described in the third aspect may be performed. The apparatus may further include a communication interface, which is used for the apparatus to communicate with another device. For example, the communication interface may be a transceiver, a circuit, a bus, a module, a pin, or another type of communication interface, and the other device may be a terminal, etc. In a possible design, the apparatus may include: a memory configured to store program instructions; a communication interface configured to receive a first reference signal from a terminal device and to transmit a first codebook to the terminal device; and a processor configured to determine a first codebook based on a received first reference signal, the first codebook being used for precoding an uplink signal.

[0060] For the specific execution process of the communication interface and the processor, please refer to the description of the third embodiment, and the details will not be described again.

[0061] According to a seventh aspect, the present disclosure further provides a computer-readable storage medium comprising instructions which, when executed on a computer, enable the computer to perform a method according to any one of the first to fourth aspects.

[0062] According to an eighth aspect, the present disclosure further provides a chip system, the chip system including a processor, and possibly further including a memory, configured to perform the method of any one of the first to fourth aspects. The chip system may include a chip, or may include a chip and another discrete component.

[0063] According to a ninth aspect, the present disclosure further provides a computer program product comprising instructions which, when executed on a computer, enable the computer to perform a method according to any one of the first to fourth aspects.

[0064] According to a tenth aspect, the present disclosure further provides a system, the system comprising an apparatus of the fifth aspect and an apparatus of the sixth aspect. [Brief description of the drawings]

[0065] [Figure 1] FIG. 1 is a diagram of a communication system according to the present disclosure. [Diagram 2] FIG. 1 is a diagram of an AI model deployment according to the present disclosure. [Diagram 3] FIG. 1 is a diagram of an AI model deployment according to the present disclosure. [Figure 4a] FIG. 1 is a diagram of a communication system architecture according to the present disclosure. [Figure 4b] FIG. 1 is a diagram of a communication system architecture according to the present disclosure. [Figure 4c] FIG. 1 is a diagram of an application of an AI model according to the present disclosure. [Diagram 5] FIG. 1 is a diagram of a neuron according to the present disclosure. [Figure 6] FIG. 2 is a diagram of a neural network according to the present disclosure. [Figure 7] FIG. 2 is a diagram of an uplink precoding method according to the present disclosure. [Figure 8] FIG. 2 is a diagram of an uplink precoding method according to the present disclosure. [Figure 9] FIG. 2 is a diagram of a codebook generation process according to the present disclosure. [Figure 10] FIG. 2 is a diagram of an uplink precoding method according to the present disclosure. [Figure 11] FIG. 2 is a diagram of subbands according to the present disclosure. [Figure 12] FIG. 1 is a diagram of subband precoding matrix generation according to the present disclosure. [Figure 13] FIG. 2 is a diagram of an uplink precoding method according to the present disclosure. [Figure 14] FIG. 2 is a diagram of an uplink precoding method according to the present disclosure. [Figure 15] FIG. 1 is a diagram of training a CSI compression model and a precoding model according to the present disclosure. [Figure 16] FIG. 1 is a diagram of application of a CSI compression model and a precoding model in accordance with the present disclosure. [Figure 17] FIG. 2 is a diagram of a structure of a CSI compression model and a precoding model according to the present disclosure. [Figure 18] FIG. 1 is a diagram of the structure of an apparatus according to the present disclosure. [Figure 19] FIG. 1 is a diagram of the structure of an apparatus according to the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0066] FIG. 1 is an architecture diagram of a communication system 1000 to which the present disclosure can be applied. As shown in FIG. 1, the communication device includes a radio access network 100 and a core network 200. Optionally, the communication system 1000 may further include an Internet 300. The radio access network 100 may include at least one access network device (e.g., 110a and 110b in FIG. 1) and may further include at least one terminal device (e.g., 120a to 120j in FIG. 1). The terminal device is connected to the access network device in a wireless manner, and the access network device is connected to the core network in a wireless or wired manner. The core network device and the access network device may be different physical devices independent of each other, or the functions of the core network device and the logical functions of the access network device may be integrated into the same physical device, or some functions of the core network device and some functions of the access network device may be integrated into one physical device. The terminal devices may be connected to each other in a wired or wireless manner, and the access network devices may be connected to each other in a wired or wireless manner. FIG. 1 is just a diagram. The communication system may further include other network devices, such as wireless relay devices and wireless backhaul devices not shown in FIG.

[0067] The access network device may be a base station, an evolved NodeB (eNodeB), a transmission reception point (TRP), a next generation NodeB (gNB) of a 5th generation (5G) mobile communication system, an access network device of an open radio access network (O-RAN), a next generation base station of a 6th generation (6G) mobile communication system, a base station of a future mobile communication system, or an access node of a wireless fidelity (Wi-Fi) system, etc. Alternatively, the access network device may be a module or unit that completes some functions of a base station, such as a central unit (CU), a distributed unit (DU), a central unit control plane (CU-CP) module, or a central unit user plane (CU-UP) module. The access network device may be a macro base station (e.g., 110a in FIG. 1), a micro base station or an indoor base station (e.g., 110b in FIG. 1), or a relay node or a donor node, etc. The specific technology used in the access network device and the specific device form are not limited by this disclosure.

[0068] In the present disclosure, an apparatus configured to implement the function of an access network device may be an access network device, or may be an apparatus capable of supporting an access network device to implement the function, such as a chip system, a hardware circuit, a software module, or a combination of a hardware circuit and a software module. The apparatus may be installed in an access network device, or may be combined with an access network device in use. In the present disclosure, a chip system may include a chip, or may include a chip and another individual component. For ease of explanation, the following uses an example in which an apparatus configured to implement the function of an access network device is an access network device, and the access network device is a base station to describe the technical solution provided in the present disclosure.

[0069] (1) Protocol Layer Structure

[0070] The communication between the access network device and the terminal device complies with a specified protocol layer structure. The protocol layer structure may include a control plane protocol layer structure and a user plane protocol layer structure. For example, the control plane protocol layer structure may include the functions of protocol layers such as a radio resource control (RRC) layer, a packet data convergence protocol (PDCP) layer, a radio link control (RLC) layer, a media access control (MAC) layer, and a physical layer. For example, the user plane protocol layer structure may include the functions of protocol layers such as a PDCP layer, an RLC layer, a MAC layer, and a physical layer. In a possible implementation, a service data adaptation protocol (SDAP) layer may be further included above the PDCP layer.

[0071] Optionally, the protocol layer structure between the access network device and the terminal device may further include an artificial intelligence (AI) layer used to transmit data related to AI functions.

[0072] (2) Central unit (CU) and distributed unit (DU)

[0073] The access device may include a CU and a DU. A plurality of DUs may be controlled by one CU in a centralized manner. For example, an interface between a CU and a DU may be called an F1 interface. A control plane (CP) interface may be F1-C, and a user plane (UP) interface may be F1-U. The specific names of the interfaces are not limited in this disclosure. The CU and the DU may be divided based on the protocol layers of a wireless network. For example, the functions of the PDCP layer and the protocol layers higher than the PDCP layer are configured in the CU, and the functions of the protocol layers lower than the PDCP layer (e.g., the RLC layer and the MAC layer) are configured in the DU. In another example, the functions of the protocol layers higher than the PDCP layer are configured in the CU, and the functions of the PDCP layer and the protocol layers lower than the PDCP layer are configured in the DU. This is not limited.

[0074] The division of the processing functions of the CU and the DU based on the protocol layer is merely an example, and the processing functions of the CU and the DU may alternatively be divided in another manner. For example, the CU or the DU may be divided into functions having more protocol layers. In another example, the CU or the DU may be further divided into several processing functions having protocol layers. In one design, some functions of the RLC layer and functions of the protocol layers above the RLC layer are set to the CU, and the remaining functions of the RLC layer and functions of the protocol layers below the RLC layer are set to the DU. In another design, the division of the functions of the CU or the DU may alternatively be performed based on the service type or other system requirements. For example, the division may be performed based on latency. Functions whose processing time needs to meet the latency requirement are set to the DU, and functions whose processing time does not need to meet the latency requirement are set to the CU. In another design, the CU may alternatively have one or more functions of the core network. For example, the CU may be located on the network side to facilitate centralized management. In another design, a radio unit (RU) of the DU is located remotely. Optionally, the RU may have a radio frequency function.

[0075] Optionally, the DU and the RU may be divided by a physical layer (PHY). For example, the DU may implement the upper layer functions of the PHY layer, and the RU may implement the lower layer functions of the PHY layer. When the PHY layer is used for transmission, the PHY layer may include at least one of the following functions: cyclic redundancy check (CRC) code addition, channel coding, rate matching, scrambling, modulation, layer mapping, precoding, resource mapping, physical antenna mapping, or radio frequency transmission. When the PHY layer is used for reception, the PHY layer may include at least one of the following functions: CRC, channel decoding, rate matching, descrambling, demodulation, layer demapping, channel detection, resource demapping, physical antenna demapping, or radio frequency reception. The upper layer functions of the PHY layer may include some functions of the PHY layer. For example, some of these functions are closer to the MAC layer. The lower layer functions of the PHY layer may include some other functions of the PHY layer. For example, some of these functions are closer to the radio frequency functions. For example, the upper layer functions of the PHY layer may include CRC code addition, channel coding, rate matching, scrambling, modulation, and layer mapping, and the lower layer functions of the PHY layer may include precoding, resource mapping, physical antenna mapping, and radio frequency transmission functions. Alternatively, the upper layer functions of the PHY layer may include CRC code addition, channel coding, rate matching, scrambling, modulation, layer mapping, and precoding, and the lower layer functions of the PHY layer may include resource mapping, physical antenna mapping, and radio frequency transmission functions. For example, the upper layer functions of the PHY layer may include CRC, channel decoding, rate matching, decoding, demodulation, and layer de-mapping, and the lower layer functions of the PHY layer may include channel detection, resource de-mapping, physical antenna de-mapping, and radio frequency reception functions.Alternatively, the upper layer functions of the PHY layer may include CRC, channel decoding, rate de-matching, decoding, demodulation, layer de-mapping, and channel detection, and the lower layer functions of the PHY layer may include resource de-mapping, physical antenna de-mapping, and radio frequency reception functions.

[0076] For example, the functionality of the CU may be implemented by one entity or by different entities. For example, the functionality of the CU may be further divided, i.e., the control plane and the user plane are implemented separately by different entities, a control plane CU entity (i.e., a CU-CP entity) and a user plane CU entity (i.e., a CU-UP entity). The CU-CP entity and the CU-UP entity are coupled to the DU and cooperate to complete the functionality of the access network device.

[0077] Optionally, any one of the DU, CU, CU-CP, CU-UP, and RU may be a software module, a hardware structure, or a combination of a software module and a hardware structure. This is not limited. Different entities may exist in different forms. This is not limited. For example, the DU, CU, CU-CP, and CU-UP are software modules, and the RU is a hardware structure. These modules and the methods performed by these modules are also included in the protection scope of the present disclosure.

[0078] In a possible implementation, the access network device includes a CU-CP, a CU-UP, a DU, and a RU. For example, the present disclosure is performed by a DU, or by a DU and a RU, or by a CU-CP, a DU, and a RU, or by a CU-UP, a DU, and a RU. This is not limited. Methods performed by these modules are also included in the scope of protection of the present disclosure.

[0079] The terminal device may alternatively be referred to as a terminal, user equipment (UE), mobile station, or mobile terminal device. The terminal device may be widely used for communication in various scenarios, including, but not limited to, at least one of the following scenarios: device-to-device (D2D), vehicle-to-everything (V2X), machine-type communication (MTC), internet of things (IOT), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grid, smart furniture, smart office, smart wearable, smart transportation, and smart city. The terminal device may be a mobile phone, a tablet computer, a computer with wireless transceiver capability, a wearable device, a vehicle, an unmanned aerial vehicle, a helicopter, an airplane, a ship, a robot, a robot arm, or a smart home device. The specific technology and the specific device form used in the terminal device are not limited by this disclosure.

[0080] In the present disclosure, the apparatus configured to implement the function of the terminal device may be a terminal device, or may be an apparatus capable of supporting the terminal device to implement the function, such as a chip system, a hardware circuit, a software module, or a combination of a hardware circuit and a software module. The apparatus may be installed in the terminal device, or may be combined with the terminal device in use. For ease of explanation, the following will use an example in which the apparatus configured to implement the function of the terminal device is a terminal device, and the terminal device is a UE, to describe the technical solution provided in the present disclosure.

[0081] The base station and the terminal device may be fixed or mobile. The base station and / or the terminal device may be deployed on land, such as indoor or outdoor devices and handheld or vehicle-mounted devices, on water, or on airborne aircraft, balloons and satellites. The application scenarios of the base station and the terminal device are not limited in this disclosure. The base station and the terminal device may be deployed in the same scenario or in different scenarios. For example, both the base station and the terminal device are deployed on land, or the base station is deployed on land and the terminal device is deployed on water. Examples will not be repeated one by one.

[0082] The roles of base station and terminal device may be relative. For example, helicopter or unmanned aerial vehicle 120i in FIG. 1 may be configured as a mobile base station. For terminal 120j that accesses wireless access network 100 via 120i, terminal device 120i is a base station. However, for base station 110a, 120i is a terminal. In other words, 110a and 120i communicate with each other based on a wireless air interface protocol. Alternatively, 110a and 120i may communicate with each other based on an interface protocol between base stations. In this case, for 110a, 120i is also a base station. Therefore, both base station and terminal device may be collectively referred to as communication devices. 110a and 110b in FIG. 1 may be referred to as communication devices having the function of a base station, and 120a to 120j in FIG. 1 may be referred to as communication devices having the function of a terminal device.

[0083] In the present disclosure, a base station transmits a downlink signal or information to a terminal device, where the downlink information is carried on a downlink channel, and the terminal device transmits an uplink signal or information to the base station, where the uplink information is carried on an uplink channel. To communicate with the base station, the terminal device needs to establish a wireless connection with a cell controlled by the base station. The cell that establishes a wireless connection with the terminal device is called the serving cell of the terminal device. When communicating with the serving cell, the terminal device may be further interfered with by signals from neighboring cells.

[0084] In the embodiments of the present disclosure, the physical downlink shared channel (PDSCH), the physical downlink control channel (PDCCH), the physical uplink shared channel (PUSCH), and the physical uplink control channel (PUCCH) are merely examples of the downlink data channel, the downlink control channel, the uplink data channel, and the uplink control channel. In different systems and different scenarios, the data channel and the control channel may have different names. This is not limited in the embodiments of the present disclosure.

[0085] In the present disclosure, an independent network element, such as an AI network element or AI node, may be introduced into the communication system shown in FIG. 1 to perform AI-related operations. The AI-related operations include at least one of data collection, model training, model information release, model inference, or inference result release. The AI ​​network element may be directly connected to an access network device in the communication system, or indirectly connected to the access network device using a third party network element. The third party network element may be a core network element, such as an authentication management function (AMF) or a user plane function (UPF). Alternatively, the AI ​​function, AI module, or AI entity may be configured to perform AI-related operations in another network element in the communication system. For example, the other network element may be an access network device (such as a gNB), a core network device, or an operation, administration and maintenance (OAM). In this case, the network element performing the AI-related operations is a network element with built-in AI functions. The OAM is configured to perform operations, administration, maintenance, etc. on the access network device and / or the core network device.

[0086] In the present disclosure, as shown in FIG. 2 or FIG. 3, the AI ​​model may be deployed in at least one of a core network device, an access network device, a terminal device, or an OAM, and the corresponding function is implemented using the AI ​​model. In the present disclosure, the AI ​​models deployed in different nodes may be the same or different. Different models include at least one of different model structure parameters, such as the number of layers and / or weight values ​​of the model, different model input parameters, or different model output parameters. Different model input parameters and / or different model output parameters may also be referred to as different model functions. Different from FIG. 2, in FIG. 3, the access network device is divided into a CU and a DU in terms of functions. Optionally, the CU and DU may be the CU and DU of the O-RAN architecture. One or more AI models may be deployed in the CU. Additionally / alternatively, one or more AI models may be deployed in the DU. Optionally, the CU in FIG. 3 may be further divided into a CU-CP and a CU-UP. Optionally, one or more AI models may be deployed in the CU-CP. Additionally / alternatively, one or more AI models may be deployed in the CU-UP. Optionally, in Figure 2 or Figure 3, OAM of the access network devices and OAM of the core network devices may be deployed separately and independently.

[0087] Optionally, FIG. 4a is an architecture of a communication system according to the present disclosure. As shown in FIG. 4a, in a first design, the access network device includes a near-real-time radio access network intelligent controller (RAN intelligent controller, RIC) module configured to perform model training and inference. For example, the near-real-time RIC may be configured to train an AI model, and inference is performed using the AI ​​model. For example, the near-real-time RIC may obtain information about the network side and / or the terminal side from at least one of the CU, the DU, and the RU, and the information may be used as training data or inference data. Optionally, the near-real-time RIC may submit the inference result to at least one of the CU, the DU, the RU, or the terminal device. Optionally, the CU and the DU may exchange the inference result. Optionally, the DU and the RU may exchange the inference result. For example, the near-real-time RIC submits the inference result to the DU, and the DU forwards the inference result to the RU.

[0088] Alternatively, in a second design, as shown in FIG. 4a, a non-real-time RIC (optionally, the non-real-time RIC may be located in the OAM or in the core network device) is included outside the access network device and configured to perform model training and inference. For example, the non-real-time RIC is configured to train an AI model, and inference is performed using the model. For example, the non-real-time RIC may obtain information about the network side and / or the terminal side from at least one of the CU, DU, or RU. The information may be used as training data or inference data, and the inference result may be submitted to at least one of the CU, DU, RU, or terminal device. Optionally, the CU and DU may exchange the inference result. Optionally, the DU and RU may exchange the inference result. For example, the non-real-time RIC submits the inference result to the DU, and the DU forwards the inference result to the RU.

[0089] Alternatively, in a third design, as shown in FIG. 4a, the access network device includes a near-real-time RIC and a non-real-time RIC is included outside the access network device (optionally, the non-real-time RIC may be located in the OAM or in the core network device). As with the second design above, the non-real-time RIC may be configured to perform model training and inference. Additionally / alternatively, as with the first design above, the near-real-time RIC may be configured to perform model training and inference. Additionally / alternatively, the non-real-time RIC may perform model training, and the near-real-time RIC may obtain AI model information from the non-real-time RIC, obtain information about the network side and / or terminal side from at least one of the CU, DU, or RU, and use this information and the AI ​​model information to obtain an inference result. Optionally, the near-real-time RIC may submit the inference result to at least one of the CU, DU, RU, or terminal device. Optionally, the CU and DU may exchange inference results. Optionally, the DU and RU may exchange inference results. For example, the near real-time RIC submits inference results to the DU, which forwards the inference results to the RU. For example, the near real-time RIC is configured to train model A and perform inference using model A. For example, the non-real-time RIC is configured to train model B and perform inference using model B. For example, the non-real-time RIC is configured to train model C and send information about model C to the near real-time RIC. The near real-time RIC performs inference using model C.

[0090] Figure 4b is another communication system architecture according to the present disclosure. Compared with Figure 4a, in Figure 4b, the CU is separated into CU-CP and CU-UP, etc.

[0091] FIG. 4c is a diagram of an AI application framework. A data source is configured to store training data and inference data. A model training node (model training host) analyzes or trains the training data provided by the data source to obtain an AI model, and deploys the AI ​​model to a model inference node (model inference host). Optionally, the model training node may further update the AI ​​model deployed to the model inference node. The model inference node may further feed back related information of the deployed model to the model training node, so that the model training node performs optimization, update, etc. on the deployed AI model.

[0092] An AI model represents a mapping relationship between the input and output of the model. Obtaining an AI model by learning with a model training node is equivalent to obtaining a mapping relationship between the input and output of the model by learning with the model training node using training data. The model inference node uses the AI ​​model to perform inference based on inference data provided by a data source, and obtains an inference result. This method may also be described as follows: the model inference node inputs inference data to the AI ​​model, and obtains an output using the AI ​​model. The output is the inference result. The inference result may indicate a configuration parameter to be used (acted upon) by an action subject, and / or an operation to be performed by the action subject. The inference result may be planned in a unified manner by an actor entity and sent to one or more action subjects (e.g., network entities) for action.

[0093] In this disclosure, an AI model implements a particular method of AI function. An AI model may be a neural network or another machine learning model. An AI model may also be called a model for short. A neural network is used as an example. A neural network is a specific implementation of machine learning technology. According to the universality theorem, a neural network can theoretically approximate any continuous function, so that the neural network can learn any mapping. Traditional communication systems require extensive expertise to design communication modules. However, a neural network-based deep learning communication system can automatically discover implicit pattern structures from a large amount of data set, establish mapping relationships between data, and obtain better performance than traditional modeling methods.

[0094] The concept of neural networks comes from the neuronal structure of the brain tissue. Each neuron performs a weighted summation operation on its inputs and outputs the weighted summation result via an activation function. Figure 5 shows the structure of a neuron. The inputs of a neuron are x=[x0,x1,…,x n ], and the weight values ​​corresponding to the inputs are w=[w,w1,…,w n ] and the offset of the weighted sum is b. There are many different forms of activation functions. If a neuron's activation function is y = f(z) = max(0,z), then the output of that neuron is

[0095]

number

[0096] As another example, if the activation function of a neuron is y=f(z)=z, then the output of that neuron is

[0097]

number

[0098] Herein, b can take on a variety of possible values, such as a fractional number, an integer (including 0, a positive integer, or a negative integer, etc.), or a complex number. The activation functions of different neurons in a neural network can be the same or different.

[0099] A neural network generally includes a multi-layer structure, and each layer may include one or more neurons. Increasing the depth and / or width of a neural network can improve the representation capability of the neural network and provide a more powerful information extraction and abstract modeling capability for complex systems. The depth of a neural network may refer to the number of layers included in the neural network, and the number of neurons included in each layer may be referred to as the layer width. FIG. 6 is a diagram of the layer relationship of a neural network. In one implementation, the neural network includes an input layer and an output layer. The input layer of the neural network performs neuron processing on the received input and then transfers the result to the output layer, and the output layer obtains the output result of the neural network. In another implementation, the neural network includes an input layer, a hidden layer, and an output layer. The input layer of the neural network performs neuron processing on the received input and then transfers the result to an intermediate hidden layer. The hidden layer then transfers the calculation result to the output layer or an adjacent hidden layer. Finally, the output layer obtains the output result of the neural network. The neural network may include one hidden layer or multiple hidden layers connected in series. This is not limited. In the training process of the neural network, a loss function may be defined. The loss function describes the gap or difference between the output value of the neural network and an ideal target value. The specific form of the loss function is not limited in this disclosure. The training process of the neural network is a process of adjusting neural network parameters such as the number and width of layers of the neural network, the weight values ​​of neurons, and / or the parameters of the activation functions of neurons, so that the value of the loss function is below a threshold or meets a target requirement.

[0100] The neural network includes a deep neural network (DNN). Based on the network construction style, the DNN can be classified into a feedforward neural network (FNN), a convolutional neural network (CNN), a recurrent neural network (RNN), etc. In the present disclosure, the AI ​​model can be an FNN, a CNN, an RNN, etc., or another type of neural network, etc. This is not limited.

[0101] In a wireless communication system, there are uplink communication links and downlink communication links. The link where a terminal device transmits an uplink signal or uplink information to a base station is called an uplink communication link, and the link where a base station transmits a downlink signal or downlink information to a terminal device is called a downlink communication link. Over the past decade, most of the attention has been paid to the downlink communication link, since most of the data in wireless networks is downlink communication data. In recent years, with the emergence of services such as live broadcast and high definition video telephony, the need for uplink communication has become more and more urgent. How to improve the uplink communication rate is the current direction of research.

[0102] Uplink precoding technology is an important method to improve the uplink communication rate. In uplink precoding technology, multiple antennas (or called antenna ports) of the UE are used to assign respective weight values ​​to the data transmitted by each antenna to realize the beamforming effect, so that the signal transmitted by the UE can better match the channel condition between the UE and the base station. For example, if the signal transmitted by the terminal device is X, the uplink channel is H, the uplink precoding matrix is ​​W, and the noise is N0, then the uplink signal transmitted by the UE is S=WX, and the uplink signal received by the base station is Y=HWX+N0.

[0103] Since the precoding matrix W is expected to match the uplink channel H as best as possible, the base station first estimates the uplink channel H, calculates the corresponding precoding matrix W, and then signals W to the UE for uplink precoding. The format of W is usually a complex array, i.e. w = [w1, w2, …, w N ] T It is.

[0104] In this specification, N denotes the number of antennas of a UE, and wi denotes the weight value assigned to the i-th antenna, where wi is a floating-point complex number and the value of i ranges from 1 to N. When the base station informs the UE of the exact W, a large amount of air interface overhead needs to be consumed. Moreover, when the number of UEs increases, the feedback overhead grows exponentially. This seriously affects the air interface efficiency of the wireless network. Therefore, how to inform the UE of the precoding matrix W with low air interface overhead is an urgent problem to be solved.

[0105] In a possible implementation, a codebook-based uplink transmission may be defined, the principle of which is as follows: a set of fixed precoding codebooks (codebooks, CB) are agreed upon, each of which is applicable to the case where the UE has different quantities of antennas. Each codebook includes several precoding matrices that are different from each other. During uplink data transmission, the base station informs the UE to select a precoding matrix in the codebook to precode the uplink signal. For example, for a UE with four antennas, the agreed fixed codebook CB=[W1, W2, ..., W4] includes four optional precoding matrices W i where W i =[w1,w2,…,w4] T It is.

[0106] In one design, as shown in FIG. 7, a procedure for a base station to indicate a precoding matrix to a UE includes the following steps.

[0107] Step 701: The UE transmits a sounding reference signal (SRS) to the base station.

[0108] Step 702: The base station estimates an uplink channel H based on measurements performed on the SRS, and selects a precoding matrix matching H from a default CB.

[0109] Step 703: The base station notifies the UE of the index of the selected precoding matrix.

[0110] Step 704: After receiving the index, the UE determines, based on a pre-coding matrix corresponding to the index, and pre-codes the uplink signal based on the pre-coding matrix.

[0111] In the above design, the base station does not inform the UE of the precoding matrix W. W is a floating-point complex matrix, and its data volume is large.

[0112] Instead, the base station notifies the UE of the index of the precoding matrix W. The index is usually a small integer and its data volume is small, so that the air interface overhead can be significantly reduced. However, in this design, the agreed codebook is a universal fixed codebook, all UEs use the same codebook, and the number of candidate precoding matrices in the codebook is small. Therefore, this design lacks flexibility. For example, in a possible design, 360 degrees is equally divided into only four parts based on a four-antenna codebook, and each precoding matrix refers to one equally divided angle. Therefore, the display precision is too low. In reality, the channel environments where different UEs are located are very different. For example, some channels mainly use direct multipath, and some channels mainly use scattering multipath. Signals have different arrival directions and channel characteristics. One fixed codebook cannot adapt to the ever-changing channel environment. In addition, the precoding matrix in the codebook is usually very different from the optimal precoding matrix, and therefore the beamforming gain effect of the precoding matrix is ​​greatly reduced. For example, for a UE with four antennas, simulations show that the precoding matrix selected from the agreed-upon codebook has a demodulation loss of about 1.2 dB compared to the optimally designed precoding matrix, indicating that the agreed-upon codebook still has a large room for improvement.

[0113] The present disclosure provides an uplink precoding method, in which an individualized codebook can be configured for each UE, so that the codebook is better adapted to the actual channel environment of the UE, thereby improving the precoding accuracy of the UE. As shown in Figure 8, the method includes at least the following steps:

[0114] Step 801: The UE transmits a first reference signal to the base station, and in response, the base station receives the first reference signal from the UE.

[0115] The first reference signal may be an SRS or another uplink reference signal, etc.

[0116] Step 802: The base station determines a first codebook based on the received first reference signal, where the first codebook is used to precode the uplink signal. Optionally, the granularity of the first codebook is not limited. For example, the codebook may be configured for the UE over the entire uplink bandwidth, i.e., the first codebook may be applicable to precoding the entire uplink bandwidth. Alternatively, for each subband of the uplink bandwidth, the base station may separately configure a codebook for the UE, etc. For example, the base station configures a first codebook for the first subband, where the first codebook is applicable to precoding the first subband, etc.

[0117] For example, the UE may transmit Z reference signals separately on Z resources. The Z resources may be distinguished in the time domain, the frequency domain, and / or the spatial domain, etc., where Z is a positive integer greater than 0. For example, the UE may transmit Z reference signals separately at Z time instants. In this case, the Z resources are distinguished in the time domain. The base station separately measures the received Z reference signals to obtain uplink channel information. The uplink channel information may be a characteristic matrix H of the uplink channel. For ease of description, the characteristic matrix H of the uplink channel may alternatively be referred to as an uplink channel H. A first codebook is determined based on the Z uplink channel information, where the first codebook may be referred to as an individualized codebook of the UE.

[0118] In one implementation, the base station may determine a first codebook based on Z pieces of uplink channel information and a codebook generation model. The codebook generation model is used to determine the first codebook based on the Z pieces of uplink channel information. The name of the codebook generation model is not limited. For example, the codebook generation model may alternatively be referred to as a model, an AI model, etc.

[0119] For example, Z uplink channel information, e.g. Z H, may be used as input of the codebook generation model, and the output of the codebook generation model is the first codebook. For example, as shown in FIG. 9, uplink channel information h1 corresponding to UE1 is input to the codebook generation model, and the output is the codebook corresponding to UE1, and uplink channel information h2 corresponding to UE2 is input to the codebook generation model, and the output is the codebook corresponding to UE2. Alternatively, processing may be performed on the Z uplink channel information. The processing may include, but is not limited to, quantizing the Z H, separating the real part from the imaginary part, or separating the phase from the amplitude, etc. The processed Z H are used as input of the codebook generation model. Additionally / alternatively, the output of the codebook generation model is information related to the first codebook, and the first codebook is determined based on the output of the codebook generation model. For example, the first codebook may include Z precoding matrices, or Z quantized precoding matrices, or M precoding matrices, or M quantized precoding matrices, where the value of M is a positive integer less than or equal to Z and greater than 0. For example, the output of the codebook generation model is Z precoding matrices. The base station clusters the Z precoding matrices to generate M precoding matrices, and quantizes the M precoding matrices, where the first codebook includes the M quantized precoding matrices, etc. Of course, alternatively, the output of the codebook generation model may be directly designed to be the M quantized precoding matrices.

[0120] It should be noted that the above description uses an example in which the base station determines the first codebook for a single UE based on Z channel information of the single UE. In another solution, multiple UEs are configured to transmit uplink reference signals on the same resource. In this case, the uplink channel information corresponding to the UE may be obtained based on the uplink reference signals transmitted by each UE. The base station may determine a codebook applicable to multiple UEs based on the uplink channel information corresponding to multiple UEs and a codebook generation model. For example, the base station may simultaneously serve two UEs, and the two UEs transmit reference signals on the same resource. For the reference signal transmitted by UE1, the base station determines uplink channel information h1, and for the reference signal transmitted by UE2, the base station determines uplink channel information h2. In this case, the base station may use the cascaded H={h1, h2} of the channels of the two UEs as the input of the codebook generation model, and output a codebook shared by UE1 and UE2. Alternatively, the base station processes the cascaded H of the channels of the multiple UEs, and the processed information is used as an input of the codebook generation model, and / or the output of the codebook generation model is information other than the shared codebook of the multiple UEs. The base station further processes other information output by the codebook generation model to determine the codebook shared by the multiple UEs, etc.

[0121] In another implementation, the base station may determine Z precoding matrices based on the Z uplink channel information. The base station calculates the Z precoding matrices based on the Z uplink channel information H. The calculation method may include a weighted minimum mean square error (WMMSE) method, a zero forcing (ZF) method, or another method, etc. The base station may determine a first codebook based on the Z precoding matrices.

[0122] The first codebook includes Z precoding matrices, and the precoding matrices in the codebook may alternatively be referred to as codewords. Further, the Z precoding matrices may be quantized, and the first codebook includes the Z quantized precoding matrices. Alternatively, M precoding matrices are determined based on the Z precoding matrices and a precoding matrix generation model, where M is less than or equal to Z. The precoding matrix generation model is used to converge the Z precoding matrices to the M precoding matrices. In this disclosure, the name of the precoding matrix generation model is not limited. For example, the precoding matrix generation model may be referred to as a model, an AI model, or the like. In this design, the Z precoding matrices may be directly used as an input of the precoding matrix generation model and input to the precoding matrix generation model. The output of the precoding matrix generation model is the first codebook. Alternatively, the Z precoding matrices are processed, and the processed Z precoding matrices are used as inputs of a precoding matrix generation model, and / or the output of the precoding matrix generation model may alternatively be other information related to the first codebook, such that the other information is processed to obtain the first codebook, etc. The first codebook includes a small number of precoding matrices. Thus, when the base station subsequently indicates the index of the precoding matrix in the first codebook to the UE, a small signaling overhead is required. Alternatively, in addition to the above AI method, the Z precoding matrices may be clustered into M precoding matrices using a clustering method. The clustering method includes, but is not limited to, K-means, bidirectional K-means, and hierarchical clustering methods, etc. Furthermore, the M precoding matrices may be quantized to obtain M quantized precoding matrices, in which case the first codebook includes M quantized precoding matrices.

[0123] It should be noted that the above description uses an example in which a base station obtains a precoding matrix of a single UE by measuring an uplink reference signal of the single UE, and determines a first codebook of the single UE based on the precoding matrix of the single UE. In one design, when a base station serves multiple UEs simultaneously and the multiple UEs occupy the same resource, the base station may determine uplink channel information corresponding to each UE based on the reference signal transmitted by the UE. Furthermore, the precoding matrix corresponding to each UE is determined based on the uplink channel information corresponding to each UE. The codebook applicable to the multiple UEs is determined based on the precoding matrix corresponding to the multiple UEs, etc. For example, a system includes two UEs. The two UEs occupy the same resource, and the quantity of the same resource is Z. In this case, Z1 precoding matrices are determined for the reference signal transmitted by UE1 on Z resources. Z2 precoding matrices are determined for the reference signal transmitted by UE2 on Z resources. In one design, the codebook applicable to both UE1 and UE2 includes (Z1+Z2) precoding matrices, or the codebook applicable to both UE1 and UE2 includes quantized (Z1+Z2) precoding matrices. Alternatively, the base station may converge the (Z1+Z2) precoding matrices to obtain M precoding matrices. The codebook applicable to both UE1 and UE2 includes M precoding matrices, quantized M precoding matrices, etc.

[0124] Alternatively, the process of step 802 may also be described as follows: the base station estimates an uplink channel based on a reference signal transmitted by the UE, and calculates a precoding matrix that fits the uplink channel. Optionally, the range of values ​​of the calculated precoding matrix W is a complex matrix in a continuous space. The first codebook is determined based on multiple precoding matrices W calculated at multiple instants by a single UE. The quantity of the multiple precoding matrices is Z. For the process of determining the first codebook based on Z precoding matrices, please refer to the above description.

[0125] Step 803: The base station sends a first codebook to the UE, and in response, the UE receives the first codebook from the base station.

[0126] According to the above method, the base station may configure a personalized codebook for the UE. The personalized codebook may better fit the channel characteristics of the UE. The beams corresponding to the precoding matrices of the personalized codebook are no longer evenly divided in space, but may be finely divided in the main arrival direction, thereby improving the precoding accuracy of the UE.

[0127] It should be noted that in the above description, the base station may need to deploy a codebook generation model or a precoding matrix generation model. In the present disclosure, any model in the base station may be trained offline or online by the base station, or may be trained by another third party node and then transmitted to the base station. This is not limited. In the training process of the codebook generation model, supervised learning is used as an example. The training data includes training samples and labels. The training samples are Z uplink channel information. The labels are codebooks, and the codebooks are M quantized precoding matrices, etc. In this case, the model training node may determine the output of the codebook generation model based on the Z uplink channel information and the codebook generation model, and determine the value of the loss function based on the output and the label of the codebook generation model. If the value of the loss function is less than (or equal to) a threshold value or meets a target requirement, the training is completed. In the present disclosure, the value of the loss function is less than (or equal to) the threshold or meets the target requirement includes that, for one or more training samples, the value of the loss function of all training samples is less than (or equal to) the threshold or meets the target requirement, or that the weighted sum value, such as the average value, of the loss function of all training samples is less than (or equal to) the threshold or meets the target requirement. If not, the parameters of the codebook generation model are adjusted, and model training continues. If the codebook generation model is a neural network, any one of the following parameters of the model, namely, the number of layers of the neural network, the width of the neural network, the connection relationship between layers, the weight value of the neuron, the activation function of the neuron, or the offset of the activation function, etc., can be adjusted in the training process, so that the difference between the output of the neural network element and the ideal target value is as small as possible. The training process of the precoding matrix generation model is similar to that of the codebook generation model. Details will not be repeated. In the above description, supervised learning is used as an example for explanation. This is not limited.For example, the training process may further involve unsupervised learning or reinforcement learning.

[0128] Optionally, in addition to steps 801 to 803, the procedure shown in FIG. 8 may further include the following steps:

[0129] Step 804a: The UE transmits a second reference signal to the base station, and in response, the base station receives the second reference signal from the UE. The second reference signal may be an SRS or another uplink reference signal, etc. This is not limited.

[0130] Step 805a: The base station determines a first precoding matrix in a first codebook based on the received second reference signal.

[0131] For example, the base station determines an uplink channel H based on the received reference signal. The base station determines a first precoding matrix in a first codebook based on the uplink channel H. For example, the first codebook includes n precoding matrices. For the i-th precoding matrix, i is a positive integer greater than or equal to 1 and less than or equal to n. When the base station uses the i-th precoding matrix to perform analog calculations for precoding on the uplink channel H, the base station responds with an uplink transmission indicator. The uplink transmission indicator includes at least one of an uplink signal-to-interference and noise ratio or an uplink throughput, etc. The base station selects a precoding matrix with an optimal uplink transmission indicator from the n precoding matrices included in the first codebook as a first precoding matrix to be notified to the UE. For example, a precoding matrix with a maximum signal-to-interference and noise ratio is selected as the first precoding matrix, or a precoding matrix with a maximum uplink throughput is selected as the first precoding matrix.

[0132] For example, in one implementation, the base station determines a first precoding matrix corresponding to the i-th UE based on the uplink channel Hi of the i-th UE estimated at the current moment. Similar to the above description, the first codebook is set to include n precoding matrices. For each precoding matrix of the n precoding matrices, the uplink throughput corresponding to the precoding matrix when the precoding matrix precodes the uplink channel Hi is calculated. In the first codebook, the precoding matrix with the largest uplink throughput is selected as the first precoding matrix to be notified to the UE. For example, a possible formula for the base station to calculate the uplink throughput by simulation is as follows:

[0133]

number

[0134] In this specification, |H i P i | represents the transmitted signal power of the precoded uplink signal of the i-th UE, and |H k P k | represents the interference power caused by UEs other than the i-th UE to the i-th UE, and σ 2 represents the noise power,

[0135]

number

[0136] represents the signal-to-interference-and-noise ratio, or SINR, of the i-th UE, and the uplink throughput of the i-th UE can be determined by taking the logarithm of the SINR of the i-th UE (i.e., performing an lg() operation on the SINR of the i-th UE).

[0137] Step 806a: The base station sends an index of a first precoding matrix to the UE, where the index of the first precoding matrix may also be referred to as a first index, and correspondingly, the UE receives the index of the first precoding matrix from the base station.

[0138] In the present disclosure, upon receiving a first precoding matrix index, i.e., a first index, the UE may determine a first precoding matrix P corresponding to the first index based on a first codebook, and precode an uplink signal using the precoding matrix P. For example, a precoded signal may be represented as S=PX, where X represents non-precoded information and S represents the precoded signal that is actually transmitted.

[0139] Step 807a: The UE transmits a precoded uplink signal to the base station, and correspondingly, the base station receives the precoded uplink signal from the UE.

[0140] According to the above description, the base station pre-configures a personalized codebook for the UE, and when the UE actually performs uplink transmission, the base station indicates the precoding matrix in the personalized codebook to the UE.Compared with the method of configuring a fixed codebook for the UE, this method can improve precoding accuracy and channel matching.In addition, since the personalized codebook is configured for the UE, multi-user interference can be avoided to a certain extent.

[0141] In the above description of FIG. 8, the base station indicates only one precoding matrix to the UE in the entire bandwidth. The UE precodes the uplink signal corresponding to the bandwidth with the precoding matrix indicated by the base station. In this case, the UE precodes the uplink signal in the entire bandwidth with the same precoding matrix. Due to the frequency selectivity of the channel, the channel of each subband is usually different. The present disclosure provides a solution in which the base station indicates the precoding matrix to the UE in units of subbands included in the bandwidth. For example, if the entire bandwidth includes X subbands, the base station may separately indicate the corresponding precoding matrix for each of the X subbands, where X is a positive integer greater than 0. As shown in FIG. 10, a flow chart of an uplink precoding method is provided. In addition to the above steps 801 to 803, optionally, the procedure may further include the following steps.

[0142] Step 804b: The UE transmits a second reference signal to the base station, and in response, the base station receives the second reference signal from the UE.

[0143] Step 805b: The base station selects a precoding matrix in the first codebook based on the received second reference signal.

[0144] The present disclosure may include a base station determining an uplink channel H corresponding to each subband in the uplink bandwidth of the UE based on the received second reference signal, and determining a precoding matrix corresponding to each subband based on H corresponding to each subband. For example, for any one of the at least one subband included in the uplink bandwidth, the first codebook may be set as including n precoding matrices, and the base station may perform a simulation calculation to obtain a corresponding uplink signal-to-interference-and-noise ratio or a corresponding uplink throughput when H corresponding to the subband is separately precoded with each of the n precoding matrices. The base station selects a precoding matrix having a maximum uplink signal-to-interference-and-noise ratio or a maximum uplink throughput from the n precoding matrices as a precoding matrix for the subband.

[0145] Step 806b: The base station sends an index of a precoding matrix corresponding to each subband to the UE.

[0146] For example, the base station may inform the UE of the index of the precoding matrix corresponding to each subband as it is. For example, the uplink bandwidth of the UE includes two subbands, subband 1 and subband 2. The first codebook includes two precoding matrices, precoding matrix 1 and precoding matrix 2. In this case, the base station may inform the UE of the precoding matrix corresponding to subband 1, e.g., precoding matrix 1, and the precoding matrix corresponding to subband 2, e.g., precoding matrix 2.

[0147] Alternatively, due to continuous variations in the wireless channel in the frequency domain, the channels of the subbands are not exactly the same, but there is a correlation between adjacent subbands. Therefore, there is also a certain degree of correlation between the precoding matrices of adjacent subbands. In the present disclosure, the base station may use the correlation between the precoding matrices of adjacent subbands to indicate precoding matrices of different subbands to the UE. In this case, the display rule of the subband precoding matrix may be agreed upon in a protocol, or the base station may notify the UE of the display rule of the subband precoding matrix in advance. For example, in step 803, the base station notifies the UE of both the first codebook and the display rule of the subband precoding matrix. In other words, in addition to transmitting the first codebook to the UE in the above step 803, the base station should further transmit the display rule of the subband precoding matrix to the UE. For example, at least one display rule may be pre-set or pre-defined. The base station notifies the UE of the specific display rule to be used, and transmits the precoding matrix corresponding to each subband, etc. to the UE. For example, the uplink channel bandwidth includes a first subband and a second subband, and the first subband and the second subband are adjacent subbands. In this case, the display rule of the subband precoding matrix includes, for a precoding matrix corresponding to the first subband, indicating an index of the precoding matrix corresponding to the first subband, and for a precoding matrix corresponding to the second subband, indicating an offset direction and an offset value of an index of the precoding matrix of the second subband relative to an index of the precoding matrix of the first subband. The offset direction includes whether an index of the precoding matrix corresponding to the second subband increases or decreases with respect to an index of the precoding matrix corresponding to the first subband. For example, an increase may be represented by 1, and a decrease may be represented by 0. The offset value includes an absolute value of a difference between an index of the precoding matrix corresponding to the second subband and an index of the precoding matrix corresponding to the first subband.For example, if the index of the precoding matrix corresponding to the second subband is X and the index of the precoding matrix corresponding to the first subband is Y, the offset value of the two subbands is |XY|. Compared with the above method of directly indicating the precoding matrix index corresponding to each subband, this indication method can reduce the overhead for indicating the index of the precoding matrix.

[0148] For example, the first codebook configured by the base station for the UE may include 32 precoding matrices, and the index of each precoding matrix may be represented using 5 bits. Furthermore, if the entire uplink bandwidth of the UE is divided into 5 consecutive subbands, the solution of directly indicating the index of the precoding matrix corresponding to each subband requires an overhead of 5*5=25 bits for the base station to indicate the index of the precoding matrix corresponding to each subband to the UE. If the correlation between subbands is used to indicate the index variation between subbands, the overhead for indicating the precoding matrix may be reduced. Research has shown that the precoding matrix index variation between adjacent subbands usually does not exceed a certain value. For example, this value may be 4. For example, as shown in FIG. 11, if the index of the precoding matrix of subband 1 is 5, the value range of the index of the precoding matrix of subband 2 adjacent to subband 1 is from 1 to 9. In this case, the variation of the precoding matrix of subband 2 relative to the precoding matrix of subband 1 may be represented using 3 bits. One bit indicates whether the index of the precoding matrix of subband 2 increases or decreases with respect to the index of the precoding matrix of subband 1. Two bits indicate an offset value between the index of the precoding matrix of subband 2 and the index of the precoding matrix of subband 1. In an example where the entire uplink bandwidth is divided into five consecutive subbands, the precoding matrix of subband 1 needs to be indicated using five bits, and each of the precoding matrices of subband 2 to subband 5 needs to be indicated using three bits. Therefore, the precoding matrices of the five subbands need to be indicated using 5+3+3+3+3=17 bits in total. Compared with the above solution where the precoding matrices of the five subbands need to be indicated using 25 bits in total, this method can reduce the overhead for indicating the precoding matrices.

[0149] Step 807b: The UE transmits a precoded uplink signal to the base station, and correspondingly, the base station receives the precoded uplink signal from the UE.

[0150] In this disclosure, the base station indicates to the UE a precoding matrix corresponding to each subband in the entire bandwidth, and the UE precodes the uplink signal transmitted in the corresponding subband based on the corresponding precoding matrix. The above example is further used. The entire uplink bandwidth is divided into five subbands, and subband 1 to subband 5 correspond to precoding matrices P1 to P5, respectively. The UE precodes the uplink signal corresponding to subband 1 to subband 5 using P1 to P5 separately. Subband 1 is used as an example. The signal transmitted by the UE in subband 1 is represented as S=P1X1, where X1 is the non-precoded uplink signal on subband 1.

[0151] It should be noted that in the procedure of FIG. 10, the process of the base station determining the precoding matrices corresponding to different subbands may be implemented in an AI manner. For example, as shown in FIG. 12, the uplink channel h1 of UE1 is input into the precoding matrix calculation model, and the output of the model is the precoding matrix corresponding to each subband in the full-band uplink channel corresponding to UE1. In addition, the display information of the precoding matrix corresponding to each subband of UE1 is determined based on the display rule of the subband precoding matrix. In this design, the index of the precoding matrix corresponding to each subband is no longer displayed as it is, and the display rule is used to reduce the display overhead. Therefore, this process may also be referred to as a process of performing compressed display for the precoding matrix corresponding to the corresponding subband of UE1. The process of indicating the precoding matrix corresponding to each subband of UE2 is similar to the above process. The input of the precoding matrix calculation model is the uplink channel H, and the output is the precoding matrix of the subband corresponding to the uplink channel H. The name of the precoding matrix calculation model is not limited.

[0152] In this disclosure, the base station configures an individualized codebook for the UE and further supports subband-level precoding indication to improve the match between the codebook and the channel and precoding accuracy.

[0153] The present disclosure further provides an uplink precoding method, in which a base station transmits first information to a UE, and the UE independently determines a precoding matrix based on the first information and a first model. As shown in Figure 13, the method includes at least the following steps:

[0154] Step 1301: A base station determines first information based on uplink channel information and a second model.

[0155] The input of the second model is determined based on the uplink channel information, and the first information is determined based on the output of the second model. The second model may be trained offline or online by the base station, or may be obtained from a third party node. For example, the third party node is an OAM, a cloud server, or a core network device. The base station determines the uplink channel information by measuring the uplink channel. The uplink channel information is used as an input of the second model, or the uplink channel information is processed. For example, the uplink channel information is quantized, the real part of the uplink channel information is separated from the imaginary part of the uplink channel information, or the phase of the uplink channel information is separated from the amplitude of the uplink channel information, etc. The processed uplink channel information is used as an input of the second model. The output of the second model is the first information, or the output of the second model is processed to obtain the first information, etc.

[0156] Step 1302: The base station sends first information to the UE, and in response, the UE receives the first information from the base station.

[0157] Step 1303: The UE precodes an uplink signal based on a first precoding matrix.

[0158] Alternatively, step 1303 may be described as follows: the UE determines a first precoding matrix based on the first information and the first model, and the UE precodes the uplink signal based on the first precoding matrix. The first model is trained offline or online by the UE, or is from a third party node, such as a base station, an OAM, a cloud server, or a core network device. The input of the first model is determined based on the first information, and the first precoding matrix is ​​determined based on the output of the first model. For example, when receiving the first information, the UE may use the first information as an input of the first model, and input the first information directly into the first model, or may process the received first information and use the processed first information as an input of the first model, etc. The output of the first model may be the first precoding matrix. Alternatively, the output of the first model may be other information, such as the uplink channel H. The output of the first model is processed to obtain the first precoding matrix.

[0159] In the present disclosure, the granularity of the first precoding matrix determined based on the output of the first model is not limited. For example, the granularity of the first precoding matrix is ​​a bandwidth level, a subband level, a resource block (RB) level, or a resource element (RE) level. For example, if the first precoding matrix is ​​at the bandwidth level, the entire uplink bandwidth corresponds to one precoding matrix, i.e., the first precoding matrix. The UE precodes the uplink signal in the uplink bandwidth using the first precoding matrix. Alternatively, if the first precoding matrix is ​​at the subband level, one subband corresponds to one precoding matrix. The UE precodes the uplink signal of the corresponding subband using a precoding matrix corresponding to each subband, etc. Alternatively, if the first precoding matrix is ​​at the RB or RE level, each RB or RE corresponds to one precoding matrix. The UE precodes the uplink signal transmitted on each RB or RE, etc., using a precoding matrix corresponding to each RB or RE. It may be understood that the above are merely examples and are not intended to limit the present disclosure. For example, based on different designs or requirements, the first precoding matrix may alternatively be of different granularity.

[0160] Step 1304: The UE transmits a precoded uplink signal to the base station, and correspondingly, the base station receives the precoded uplink signal from the UE.

[0161] In this disclosure, an example is used in which the uplink signal is uplink data information carried on a PUSCH. The UE precodes both the PUSCH and a demodulation reference signal (DMRS) corresponding to the PUSCH using a first precoding matrix. The base station determines the first precoding matrix by estimating a received DMRS, and decodes the PUSCH based on the first precoding matrix, etc. Alternatively, the UE transmits the first precoding matrix to the base station. The base station decodes the uplink signal (e.g., a signal carried on a PUSCH) from the UE based on the received first precoding matrix, etc.

[0162] In the uplink precoding method shown in Figure 13, the UE independently determines a precoding matrix based on the first information and the first model. Since the precoding matrix is ​​no longer selected from a codebook offline, but is a floating-point matrix generated by the UE based on the first information, the precoding matrix can better adapt to the channel environment and achieve better pointing accuracy.

[0163] As shown in Figure 14, a procedure of an uplink precoding method is provided. For example, a base station sends a first model to a UE. The procedure includes at least the following steps:

[0164] Step 1401: The UE reports its capabilities to the base station, and in response, the base station receives the capabilities reported by the UE. Step 1401 is optional.

[0165] For example, when accessing a network, the UE may report its capabilities to the base station. The reported capabilities may include: Whether the UE supports running machine learning models; The type of machine learning model supported by the UE, e.g., CNN, RNN, or random forest models; The size of the memory space in the UE that is used to store the machine learning model; computing power information of the UE, which may include, for example, information such as the operating speed of the processor of the UE and / or the volume of data that can be processed by the processor, which may alternatively be referred to as the computing capability of the execution model of the UE; UE power consumption information, such as chip execution power consumption and battery capacity, or UE hardware information, such as UE antenna configuration information, e.g., the number or polarization direction of antennas and / or radio frequency channel Includes at least one of the following:

[0166] Step 1402: The base station sends information about the first model to the UE, and in response, the UE receives information about the first model from the base station.

[0167] The base station sends information about the first model to the UE only when the capabilities reported by the UE support machine learning. In this design, the base station trains the first model and the second model offline or online.

[0168] In one design, the second model is a precoding compression model, and the first model is a precoding recovery model. In this disclosure, an example in which the base station performs the model training is used for explanation. The method of performing the model training by another node is similar. Optionally, the samples used by the other node to perform the model training are from the base station or are determined based on a given channel generation algorithm or a given channel generation model. This is not limited. Supervised learning is used as an example. The training data collected by the base station includes training samples and labels. The training samples are uplink channels H, and the labels are precoding matrices corresponding to the uplink channels H. The second model is a precoding compression model. The base station uses H as an input of the second model, and the output of the second model is a compressed precoding matrix. Alternatively, the base station determines a precoding matrix corresponding to the uplink channel H in an AI manner or a non-AI manner. The precoding matrix is ​​used as an input of the second model, and the output is a compressed precoding matrix. The compressed precoding matrix is ​​used as an input of a first model, and the output is a precoding matrix. The base station compares the precoding matrix output by the first model with the label to determine a value of a loss function. Optionally, the loss function is a mean square error (MSE) function. If the value of the loss function is less than a threshold or meets a target requirement, model training is stopped. Alternatively, if the value of the loss function is equal to or greater than a threshold or does not meet a target requirement, parameters of the first model and / or parameters of the second model are adjusted, and model training continues. In this design, the second model is used to compress the precoding matrix. The second model may alternatively be referred to as a precoding compression model or have another name. The first model is used to determine or restore the precoding matrix based on the received first information. The first model may alternatively be referred to as a precoding restoration model or have another name.

[0169] In another design, the model training process is described using an example in which the second model is a CSI compression model and the first model is a precoding model.

[0170] In a possible model training method, the CSI compression model and the precoding model are jointly trained. The training process is shown in FIG. 15. The input of the CSI compression model is the uplink channel H, and the output of the CSI compression model is the compressed CSI. The compressed CSI of each UE is input to the precoding model corresponding to that UE, and the output of the precoding model is the precoding matrix of that UE, or the output of the precoding model is the uplink channel information. The precoding matrix is ​​determined based on the uplink channel information. The precoding matrices of all UEs are combined and substituted into a loss function. For the loss function, the parameters of the CSI model and the parameters of the precoding model of each UE are optimized. For example, a backpropagation method such as stochastic gradient descent may be used to obtain the optimal model parameters using an iterative algorithm. In this method, no labels are required. Therefore, this method belongs to the category of unsupervised learning. In FIG. 15, the joint training of the CSI compression model and the precoding model of three UEs is used as an example, which is not intended to limit the present disclosure.

[0171] Alternatively, the precoding model and the CSI compression model of each UE are trained separately. The training process includes: a base station collects training data, where the training data includes training samples and labels, where the training samples are uplink channels H corresponding to the UEs, and the labels are precoding matrices corresponding to the UEs; the uplink channels H are used as inputs of the CSI compression models, and the output is compressed CSI corresponding to the UEs; the compressed CSI is input to the precoding models corresponding to the UEs and output as precoding matrices corresponding to the UEs or output as uplink channel information, where the precoding matrices are determined based on the uplink channel information; a value of a loss function is calculated based on the precoding matrices and labels output by the precoding models; when the value of the loss function is less than a threshold or meets a target requirement, the model training may be terminated, or when the value of the loss function is equal to or greater than a threshold or does not meet a target requirement, parameters of at least one of the precoding models or the CSI compression models are updated or optimized, and the model training continues.

[0172] In this design, the CSI compression model is responsible for compressing the uplink channel H, outputting corresponding compressed CSI for each UE, and transmitting the compressed CSI to the UE. The UE processes the received compressed CSI to generate a precoding matrix. Because there is correlation between the subbands and the even-numbered subcarriers of H, the CSI compression model at the base station side can compress H. In addition, because precoding mainly plays a role in adapting the channel and avoiding interference, the CSI compression model can extract the main characteristics of the channel and the main characteristics of the interference, and these characteristics can be reflected in the compressed CSI. The UE may design a precoding matrix based on the main characteristics included in the compressed CSI to achieve an ideal precoding effect.

[0173] In the present disclosure, after training the first model and the second model, the base station may send information about the first model to the UE. The information about the first model includes at least one of the following: a parameter of the model, an input format of the model, or an output format of the model, etc. A neural network is used as an example. The parameters of the model include at least one of the number of layers of the neural network, the width of the neural network, the connection relationship between layers, the weight of a neuron, the activation function of a neuron, or the offset of the activation function, etc. The UE restores or determines the first model based on the information about the first model.

[0174] Step 1403: The UE transmits a reference signal to the base station, and in response, the base station receives a reference signal from the UE.

[0175] The base station performs, eg, estimation or measurement, on the received reference signals to determine uplink channel information.

[0176] Step 1404: The base station determines first information based on the uplink channel information and the second model.

[0177] Step 1405: The base station sends first information to the UE, and in response, the UE receives the first information from the base station.

[0178] Step 1406: The UE determines a precoding matrix based on the first information and the first model, and precodes the uplink signal with the precoding matrix.

[0179] Step 1407: The UE transmits a precoded uplink signal to the base station, and correspondingly, the base station receives the precoded uplink signal from the UE.

[0180] In one design, the second model is a precoding compression model, and the first model is a precoding recovery model. The base station uses uplink channel information determined based on a reference signal transmitted by the terminal as an input of the second model, or uses information obtained after the uplink channel information is processed as an input of the second model. The output of the second model is a compressed precoding matrix, or the information obtained after the output of the second model is processed is a compressed precoding matrix. Alternatively, the base station may determine a precoding matrix corresponding to the uplink channel information in an AI or non-AI manner, and use the precoding matrix as an input of the second model, or use the processed precoding matrix as an input of the second model. The output of the second model is a compressed precoding matrix. The first information may indicate a compressed precoding matrix, and the first information may explicitly or implicitly indicate the compressed precoding matrix. Thereafter, upon receiving the first information, the UE may determine a compressed precoding matrix based on the first information. The compressed precoding matrix is ​​used as an input of the first model, or the information obtained after the compressed precoding matrix is ​​processed is used as an input of the first model, the output of the first model includes the precoding matrix, or the information obtained after the information output by the first model is processed includes the precoding matrix, etc.

[0181] In another design, the second model is a CSI compression model and the first model is a precoding model. The base station uses uplink channel information as an input of the second model, and an output of the second model is compressed CSI. The first information indicates compressed CSI. The first information may explicitly or implicitly indicate compressed CSI. Upon receiving the first information, the terminal may use the compressed CSI as an input of the first model, or use information obtained after the compressed CSI is processed as an input of the first model. The output of the first model includes uplink channel information. The terminal determines a precoding matrix based on the uplink channel information. Specifically, the terminal may convert the uplink channel information into a precoding matrix, etc. in an AI or non-AI manner. The non-AI manner includes, but is not limited to, a WMMSE method, a zero forcing ZF method, or another method, etc. Alternatively, the output of the first model includes the precoding matrix, etc. as is.

[0182] It should be noted that in the above description, determining a precoding matrix of a single UE is used as an illustrative example. On the base station side, when the base station serves multiple UEs and the multiple UEs occupy the same resource, the first information of the multiple UEs may be determined simultaneously based on the above second model. For example, the second model is a CSI compression model and the first model is a precoding model.

[0183] As shown in FIG. 16, a base station serves two UEs, and the two UEs are configured to occupy the same resource. In this case, the base station measures uplink channel information corresponding to the two UEs separately. The two UEs are called h1 and h2, respectively. The base station determines a cascaded H of the channel based on h1 and h2, where H={h1, h2}. The cascaded H is used as the input of a CSI compression model, and the output is the cascaded compressed CSI, where the cascaded compressed CSI is G=[G1, G2, ..., G U ], where the value of U corresponds to the quantity of UEs. In a scenario where two UEs occupy the same resource, the value of U is 2, i.e., in this scenario, the output of the CSI compression model is G=[G1,G2], where G1 represents the compressed CSI corresponding to UE1 and G2 represents the compressed CSI corresponding to UE2. In this disclosure, the compressed CSI may be expressed as: G i = [g1,g2,…,g L ]

[0184] In this specification, i represents the number of a UE. The value range of i is a positive integer less than or equal to U, and the value range of U is a positive integer greater than or equal to 1. L indicates the length of the CSI fed back over the air interface. In addition, in order to reduce the air interface overhead for feeding back the compressed CSI, the G i The length of the piece may be interrupted.

[0185] In this disclosure, as shown in FIG. 16, the compressed CSI of UE1 is referred to as compressed CSI1, and the compressed CSI1 is configured to be G1 above. The compressed CSI1 is used as an input of a precoding model 1 corresponding to UE1, and the output is a precoding matrix W1. UE1 precodes the uplink signal of UE1 using W1. The compressed CSI corresponding to UE2 is configured to be referred to as compressed CSI2. The compressed CSI2 is used as an input of a precoding model 2 corresponding to UE2, and the output is a precoding matrix W2. UE2 precodes the uplink signal of UE2 using W2.

[0186] Alternatively, in this disclosure, the base station may compress the uplink channel H of each UE separately. For example, the base station determines a characteristic matrix H of the uplink channel corresponding to each UE, where H is used as an input of a CSI compression model, and the output of the CSI compression model is the compressed CSI corresponding to the UE. For example, the base station may use uplink channel information h1 corresponding to UE1 as an input of the CSI compression model, and the output is the compressed CSI corresponding to UE1, i.e., compressed CSI1. Alternatively, the base station may use uplink channel information h2 corresponding to UE2 as an input of the CSI compression model, and the output is the compressed CSI corresponding to UE2, i.e., compressed CSI2.

[0187] In one design, as shown in FIG. 17, the CSI compression model may include a 5-layer convolutional layer network and a 2-layer fully connected layer network, and the precoding model may include a 4-layer convolutional layer network and a 3-layer fully connected layer network, etc.

[0188] It should be noted that in this disclosure, the meaning of H is described as follows: That is, H may also be referred to as uplink channel information, uplink channel characteristic matrix, or uplink channel response, etc. In this disclosure, H may refer to the uplink channel information of a single UE, or may indicate the uplink channel information of multiple UEs. For example, a system includes three UEs, and the three UEs occupy the same resource. In this case, the base station may measure the reference signals respectively transmitted by the three UEs on the corresponding resources, and determine the uplink channel information hi corresponding to each UE, in which case H may refer to the cascaded uplink channel information H=[h1, h2, h3] of the three UEs.

[0189] In the present disclosure, the codebook is no longer used, and a second model on the base station side and a first model on the UE side are designed. For example, the second model is a compressed CSI model. In the compressed CSI model, the CSI may be compressed into a bit string with a small length. The base station may control the air interface overhead by controlling the length of the CSI, and a small air interface overhead may be realized by design. On the UE side, a precoding matrix is ​​generated using the precoding model and the compressed CSI. In this design, since the precoding matrix obtained by the UE is not selected from a separate codebook, the precoding matrix can better adapt to the channel environment and have better pointing accuracy. Furthermore, the second model is designed such that the output of the second model supports precoding matrices of various granularities, such as a bandwidth-level precoding matrix, a subband-level precoding matrix, or a RE-level precoding matrix.

[0190] Moreover, both Fig. 8 and Fig. 10 show solutions based on explicit codebooks. Compared with feeding back the original precoding floating-point matrix W, the feedback overhead can be significantly reduced by feeding back the codewords of the precoding matrix in the codebook. However, the codewords in the codebook are discrete. Therefore, there is a difference from the optimal precoding matrix W. The difference can be reduced by increasing the number of codewords in the codebook, but the larger the codebook, the larger the number of index bits will be, and the larger the feedback overhead will be. In the solutions of Fig. 13 or Fig. 14, the low overhead representation of precoding is completed using two adaptive AI models at the base station side and the UE side, respectively.

[0191] It can be understood that in order to implement the functions of the above methods, the base station and the terminal include corresponding hardware structures and / or software modules for performing these functions. Those skilled in the art shall easily realize that the present disclosure can be implemented by hardware or a combination of hardware and computer software by referring to the units and method steps in the examples described in the present disclosure. Whether the functions are performed by hardware or by hardware driven by computer software depends on the individual application scenarios and design constraints of the technical solutions.

[0192] 18 and 19 are diagrams of possible structures of communication devices according to the present disclosure, respectively. These communication devices may be configured to implement the functions of the terminal or base station in the above method, and thus may also realize the advantageous effects of the above method. In the present disclosure, the communication device may be one of the terminals 120a to 120j shown in FIG. 1, may be the base station 119a or 119b shown in FIG. 1, or may be a module (e.g., a chip) used in the terminal or base station.

[0193] As shown in FIG. 18, the communication device 1800 includes a processing unit 1810 and a transceiver unit 1820. The transceiver unit may alternatively be referred to as a communication unit. The communication device 1800 is configured to implement the functions of a terminal or a base station in the method shown in FIG. 8, FIG. 10, or FIG. 13.

[0194] When the communications apparatus 1800 is configured to implement the functions of a base station in the method shown in FIG. 8 or 10, the transceiver unit 1820 is configured to receive a first reference signal from a terminal device, the processing unit 1810 is configured to determine a first codebook based on the received first reference signal, where the first codebook is used to precode an uplink signal, and the transceiver unit 1820 is further configured to transmit the first codebook to the terminal device.

[0195] When the communications apparatus 1800 is configured to implement the functions of a UE in the method shown in FIG. 8 or FIG. 10, the transceiver unit 1820 is configured to transmit a first reference signal to an access network device and receive a first codebook from the access network device, and the processing unit 1810 is configured to precode an uplink signal using the first codebook.

[0196] When the communications apparatus 1800 is configured to implement the functionality of a UE in the method shown in FIG. 13 , the transceiver unit 1820 is configured to receive first information from the access network device, the processing unit 1810 is configured to precode an uplink signal based on a first precoding matrix, where the first precoding matrix is ​​determined based on an output of a first model and an input of the first model is determined based on the first information, and the transceiver unit 1820 is further configured to transmit the precoded uplink signal to the access network device.

[0197] When the communications apparatus 1800 is configured to implement the functions of a base station in the method shown in FIG. 13 , the processing unit 1810 is configured to determine first information based on uplink channel information and a second model, where an input of the second model is determined based on the uplink channel information and the first information is determined based on an output of the second model, and the transceiver unit 1820 is configured to transmit the first information to the terminal device, where the first information indicates a compressed first precoding matrix, or the first information indicates compressed channel state information CSI.

[0198] For further detailed description of the processing unit 1810 and the transceiver unit 1820, please directly refer to the relevant description of the method shown in Figure 8, Figure 10, or Figure 13. Details will not be described here.

[0199] As shown in FIG. 19, the communication device 1900 includes a processor 1910 and an interface circuit 1920. The processor 1910 and the interface circuit 1920 are coupled to each other. It may be understood that the interface circuit 1920 may be a transceiver or an input / output interface. Optionally, the communication device 1900 may further include a memory 1930 configured to store instructions executed by the processor 1910, to store input data required to execute the instructions by the processor 1910, or to store data generated after the processor 1910 executes the instructions.

[0200] When the communications device 1900 is configured to perform the above method, the processor 1910 is configured to implement the functions of the processing unit 1810 and the interface circuit 1920 is configured to implement the functions of the communications unit 1820.

[0201] When the communication device is a chip used in a terminal, the chip in the terminal implements the function of the terminal in the above method. The chip in the terminal receives information from another module (such as a radio frequency module or an antenna) in the terminal, and this information is transmitted to the terminal by the base station, or the chip in the terminal transmits information to another module (such as a radio frequency module or an antenna) in the terminal, and this information is transmitted to the base station by the terminal.

[0202] When the communication device is a module used in a base station, the module in the base station implements the functions of the base station in the above method. The module in the base station receives information from another module (e.g., a radio frequency module or an antenna) in the base station, which is transmitted by the terminal to the base station, or the module in the base station transmits information to another module (e.g., a radio frequency module or an antenna) in the base station, which is transmitted by the base station to the terminal. The module in the base station in this specification may be a baseband chip in the base station, or a DU or another module. The DU in this specification may be a DU in an open radio access network (O-RAN) architecture.

[0203] It may be understood that a processor in this disclosure may be a central processing unit (CPU), another general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a transistor logic device, a hardware component, or any combination thereof. A general-purpose processor may be a microprocessor or any general processor.

[0204] Memory in this disclosure may be random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disk, removable hard disk, CD-ROM, or any other form of storage medium known in the art.

[0205] For example, a storage medium may be coupled to the processor such that the processor can read information from, and write information to, the storage medium. The storage medium may alternatively be components of the processor. The processor and the storage medium may reside in an ASIC. Further, the ASIC may reside in a base station or terminal. Of course, the processor and the storage medium may reside as separate components in a base station or terminal.

[0206] All or part of the method of the present disclosure may be implemented by software, hardware, firmware, or any combination thereof. When software is used to implement the method, all or part of the method may be implemented in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer programs or instructions are loaded and executed on a computer, all or part of the procedures or functions according to the present disclosure are executed. The computer may be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user equipment, a core network device, an OAM, or another programmable device. The computer program or instructions may be stored in a computer-readable storage medium or may be transmitted from a computer-readable storage medium to another computer-readable storage medium. For example, the computer program or instructions may be transmitted from a website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless manner. The computer-readable storage medium may be any available medium that can be accessed by a computer, or a data storage device, such as a server or data center, that incorporates one or more available media. The available media may be magnetic media, such as, for example, a floppy disk, a hard disk, or a magnetic tape, optical media, such as, for example, a digital video disk, or semiconductor media, such as, for example, a solid-state disk. The computer readable storage medium may be a volatile or non-volatile storage medium or may include two types of storage media, namely, volatile storage media and non-volatile storage media.

[0207] In this disclosure, unless otherwise specified or there is no logical contradiction, the terms and / or descriptions in different embodiments are consistent and may be mutually referenced, and the technical features of different embodiments may be combined based on their internal logical relationships to form a new embodiment.

[0208] In the present disclosure, "at least one" means one or more, and "multiple" means two or more. "And / or" indicates an association relationship between related objects, and represents that three relationships may exist. For example, A and / or B may represent the following cases: A only exists, A and B both exist, and B only exists, where A and B may be singular or plural. In the description of the present disclosure, the character " / " indicates an "or" relationship between related objects. In the formula of the present disclosure, the character " / " indicates a "division" relationship between related objects. "Comprising at least one of A, B, or C" may indicate including A, including B, including C, including A and B, including A and C, including B and C, and including A, B, and C.

[0209] It can be understood that various numbers in this disclosure are used merely for distinction to facilitate description, and are not used to limit the scope of the present disclosure. The sequential numbers of the above processes do not imply an execution order, and the execution order of the processes shall be determined based on the functions and internal logic of the processes.

Claims

1. receiving first information from an access network device; Precoding an uplink signal based on a first precoding matrix, the first precoding matrix is ​​determined based on an output of a first model, and an input of the first model is determined based on the first information; and transmitting a precoded uplink signal to the access network device.

2. the first information indicates a compressed first precoding matrix and the output of the first model includes the first precoding matrix; or The first information indicates compressed channel state information (CSI), the output of the first model includes uplink channel information, and the first precoding matrix is ​​determined based on the uplink channel information; or 2. The method of claim 1, wherein the first information indicates compressed CSI and the output of the first model includes the first precoding matrix.

3. The method of claim 1 or 2, wherein the first precoding matrix is ​​at a wideband level, a subband level, or a resource element level.

4. determining first information based on uplink channel information and a second model, an input of the second model is determined based on the uplink channel information and the first information is determined based on an output of the second model; transmitting the first information to a terminal device, the first information indicates a compressed first precoding matrix, or the first information indicates compressed channel state information (CSI); An uplink precoding method comprising:

5. receiving a first reference signal from a terminal device; determining a first codebook based on the received first reference signal, the first codebook being used to precode an uplink signal; transmitting the first codebook to the terminal device; An uplink precoding method comprising:

6. The step of determining a first codebook based on the received first reference signal includes the steps of: separately measuring the first reference signal on Z resources to obtain Z uplink channel information, where Z is an integer equal to or greater than 1; and determining the first codebook based on the Z uplink channel information.

7. The step of determining the first codebook based on the Z uplink channel information comprises: The method of claim 6 , comprising determining the first codebook based on the Z uplink channel information and a codebook generation model.

8. The step of determining the first codebook based on the Z uplink channel information comprises: determining Z precoding matrices based on the Z uplink channel information; determining the first codebook based on the Z precoding matrices; 7. The method of claim 6, comprising:

9. The step of determining the first codebook based on the Z precoding matrices comprises: determining M precoding matrices based on the Z precoding matrices and a precoding matrix generation model, where M is less than or equal to Z; The method of claim 8 , comprising the step of: the first codebook comprising the M precoding matrices.

10. The step of determining the first codebook based on the Z precoding matrices comprises: determining M precoding matrices based on the Z precoding matrices and a precoding matrix generation model, where M is less than or equal to N; quantizing the M precoding matrices separately to obtain M quantized precoding matrices, wherein the first codebook includes the M quantized precoding matrices.

11. The method comprises: transmitting indication information to the terminal device, the indication information indicating X precoding matrices corresponding to X subbands, the X precoding matrices belonging to the first codebook, each of the X precoding matrices being used to precode an uplink signal on a corresponding subband, where X is a positive integer; 11. The method of any one of claims 5 to 10, further comprising:

12. The step of transmitting the display information to the terminal device includes: The method of claim 11 , comprising: transmitting the indication information to the terminal device according to an indication rule of a subband precoding matrix.

13. The method comprises: sending the representation rule of the subband precoding matrix to the terminal device; The method of claim 11 or 12 further comprising:

14. The representation rule of the subband precoding matrix is: the uplink channel bandwidth includes at least a first subband and a second subband; The indication rule of the precoding matrix corresponding to the first subband includes indicating an index of the precoding matrix corresponding to the first subband; 14. The method according to claim 12 or 13, wherein the display rule of the precoding matrix corresponding to the second subband includes a step of indicating an offset direction and an offset value of an index of the precoding matrix of the second subband relative to the index of the precoding matrix of the first subband.

15. 11. The method of claim 5, wherein the first codebook is a codebook for the first subband, and the first codebook is used to precode an uplink signal on the first subband.

16. transmitting a first reference signal to an access network device; receiving a first codebook from the access network device; and precoding an uplink signal using the first codebook.

17. The step of precoding an uplink signal using the first codebook comprises: receiving indication information from the access network device, the indication information indicating X precoding matrices corresponding to X subbands, and the X precoding matrices belonging to the first codebook; precoding an uplink signal on a corresponding subband with each of the X precoding matrices; 17. The method of claim 16, comprising:

18. The method comprises: The method according to claim 16 or 17, further comprising the step of receiving a representation rule of a subband precoding matrix from the access network device.

19. The representation rule of the subband precoding matrix is: the uplink channel bandwidth includes at least a first subband and a second subband; The indication rule of the precoding matrix corresponding to the first subband includes indicating an index of the precoding matrix corresponding to the first subband; The display rule of the precoding matrix corresponding to the second subband includes a step of indicating an offset direction and an offset value of an index of the precoding matrix of the second subband relative to the index of the precoding matrix of the first subband.

20. The method of claim 18.

20. The first codebook is a codebook of the first subband, and the step of precoding an uplink signal using the first codebook includes:

17. The method of claim 16, comprising precoding an uplink signal on the first subband with the first codebook.

21. A communication device comprising a unit arranged to carry out the method according to any one of claims 1 to 3 or a unit arranged to carry out the method according to any one of claims 16 to 20.

22. A communications device including a processor and a memory, the processor coupled to the memory and configured to perform a method according to any one of claims 1 to 3 or any one of claims 16 to 20.

23. A communication device comprising a unit adapted to carry out the method according to claim 4 or a unit adapted to carry out the method according to any one of claims 5 to 15.

24. A communications device including a processor and a memory, the processor coupled to the memory, the processor configured to perform the method of claim 4 or any one of the methods of claims 5 to 15.

25. A communication system comprising a communication device according to claim 21 or 22 and a communication device according to claim 23 or 24.

26. A computer readable storage medium storing instructions which, when executed on a computer, enable the computer to perform the method according to any one of claims 1 to 3, or the method according to claim 4, or the method according to any one of claims 5 to 15, or the method according to any one of claims 16 to 20.

27. A computer program product comprising instructions which, when executed on a computer, enable the computer to carry out the method according to any one of claims 1 to 3, or the method according to claim 4, or the method according to any one of claims 5 to 15, or the method according to any one of claims 16 to 20.

Images