Channel state information feedback method, data transmission method, device and system

Abstract

Provided in an embodiment of the present application are a data transmission method, a channel state information feedback method, device and system, the data transmission method comprising: a network device transmitting first information to a terminal device, the first information instructing the terminal device to generate and feed back channel state information; the network equipment receives N layers of channel state information sent by terminal equipment, N is greater than or equal to 2, and at least two layers of channel state information in the N layers of channel state information are indicated by unequal bit numbers; and the network equipment sends data to the terminal equipment according to the channel state information of the N layers.

Description

Channel state information feedback method, data transmission method, device and system Technical Field

[0001] The present application relates to the field of communications. Background Art

[0002] In Multiple Input Multiple Output (MIMO) technology, a terminal device measures the spatial channel and feeds back Channel State Information (CSI) to the base station. Based on the CSI reported by the terminal device, the base station selects an appropriate precoding matrix for downlink transmission to the terminal device, thereby minimizing the probability of bit errors in the terminal device's reception.

[0003] It should be noted that the above introduction to the technical background is merely intended to provide a clear and complete description of the technical solutions of this application and facilitate understanding by those skilled in the art. Simply because these solutions are described in the background technology section of this application, it should not be assumed that the above technical solutions are well known to those skilled in the art.

[0004] Summary of the Invention

[0005] The inventors have discovered that, since the prior art uses the same modulation coding scheme (MCS) for different downlink transmission layers, the design of the MCS should satisfy the reliable transmission of the sub-channel (layer) with the worst signal-to-noise ratio.

[0006] To address the above issues, embodiments of the present application provide a channel state information feedback method, data transmission method, device, and system, which improve the transmission performance of all sub-channels by reinforcing one or more relatively poor channels.

[0007] According to one aspect of an embodiment of the present application, a channel state information feedback apparatus is provided, configured in a terminal device, the apparatus comprising:

[0008] a receiving unit configured to receive first information sent by a network device, where the first information instructs the terminal device to generate and feed back channel state information;

[0009] a generating unit configured to generate N layers of channel state information according to a first rule that is predefined or preconfigured or configured by the network device, where N ≥ 2, wherein channel state information of at least two layers of the N layers of channel state information is indicated by unequal numbers of bits;

[0010] A sending unit is configured to send the channel state information of the N layer to the network device.

[0011] According to another aspect of an embodiment of the present application, there is provided a data sending device, configured in a network device, the device comprising:

[0012] A first sending unit, configured to send first information to a terminal device, where the first information instructs the terminal device to generate and feed back channel state information;

[0013] a receiving unit configured to receive N layers of channel state information sent by a terminal device, where N≥2, and channel state information of at least two layers of the N layers of channel state information is indicated by unequal numbers of bits;

[0014] A second sending unit sends data to the terminal device according to the channel state information of the N layer.

[0015] One of the beneficial effects of the embodiments of the present application is that, according to the embodiments of the present application, the transmission performance of all sub-channels is improved by reinforcing one or more relatively poor channels.

[0016] With reference to the following description and accompanying drawings, specific embodiments of the present application are disclosed in detail, indicating the manner in which the principles of the present application can be employed. It should be understood that the embodiments of the present application are not limited in scope. Within the spirit and scope of the appended claims, the embodiments of the present application include many variations, modifications and equivalents.

[0017] Features described and / or illustrated with respect to one embodiment may be used in the same or similar manner in one or more other embodiments, combined with features in other embodiments, or substituted for features in other embodiments.

[0018] It should be emphasized that the term "include / comprising" when used herein refers to the presence of features, integers, steps or components, but does not exclude the presence or addition of one or more other features, integers, steps or components. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The elements and features described in one figure or one embodiment of the present application can be combined with the elements and features shown in one or more other figures or embodiments. In addition, in the accompanying drawings, similar reference numerals represent corresponding parts in several figures and can be used to indicate corresponding parts used in more than one embodiment.

[0020] The included drawings are used to provide a further understanding of the embodiments of the present application, which constitute a part of the specification, are used to illustrate the implementation methods of the present application, and together with the text description, explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without inventive work. In the drawings:

[0021] FIG1 is a schematic diagram of generating and feeding back CSI based on a traditional codebook method;

[0022] FIG2 is a schematic diagram of generating and feeding back CSI based on AI / ML methods;

[0023] Figure 3 is a schematic diagram of a bilateral AI / ML model;

[0024] FIG4 is a schematic diagram of a method for feeding back channel state information according to an embodiment of the present application;

[0025] FIG5 is a schematic diagram of an example of a method according to an embodiment of the present application;

[0026] FIG6 is a schematic diagram of another example of a method according to an embodiment of the present application;

[0027] FIG7 is a schematic diagram of yet another example of a method according to an embodiment of the present application;

[0028] FIG8 is a schematic diagram of a data sending method according to an embodiment of the present application;

[0029] FIG9 is a schematic diagram of a channel state information feedback device according to an embodiment of the present application;

[0030] FIG10 is a schematic diagram of a data transmitting device according to the present embodiment;

[0031] FIG11 is a schematic diagram of a communication system according to an embodiment of the present application;

[0032] FIG12 is a schematic diagram of a terminal device according to an embodiment of the present application;

[0033] FIG13 is a schematic diagram of a network device according to an embodiment of the present application. DETAILED DESCRIPTION

[0034] The above and other features of the present application will become apparent through the following description with reference to the accompanying drawings. In the description and the accompanying drawings, specific embodiments of the present application are disclosed in detail, which illustrate some embodiments in which the principles of the present application can be adopted. It should be understood that the present application is not limited to the described embodiments. On the contrary, the present application includes all modifications, variations and equivalents that fall within the scope of the appended claims.

[0035] In the embodiments of the present application, the terms "first", "second", etc. are used to distinguish different elements from the name, but do not indicate the spatial arrangement or temporal order of these elements, and these elements should not be limited by these terms. The term "and / or" includes any one and all combinations of one or more of the associated listed terms. The terms "comprising", "including", "having", etc. refer to the presence of the stated features, elements, components or components, but do not exclude the presence or addition of one or more other features, elements, components or components.

[0036] In the embodiments of this application, the singular forms "a," "the," etc. include plural forms and should be broadly understood to mean "a" or "a type" rather than being limited to "one." Furthermore, the term "said" should be understood to include both singular and plural forms, unless the context clearly indicates otherwise. Furthermore, the term "according to" should be understood to mean "at least in part based on...", and the term "based on" should be understood to mean "at least in part based on...", unless the context clearly indicates otherwise.

[0037] In the embodiments of the present application, the term "communication network" or "wireless communication network" may refer to a network that complies with any of the following communication standards, such as Long Term Evolution (LTE), enhanced Long Term Evolution (LTE-A, LTE-Advanced), Wideband Code Division Multiple Access (WCDMA), High-Speed ​​Packet Access (HSPA), etc.

[0038] Furthermore, communication between devices in the communication system may be carried out according to communication protocols of any stage, for example, including but not limited to the following communication protocols: 1G (generation), 2G, 2.5G, 2.75G, 3G, 4G, 4.5G and future 5G, New Radio (NR), etc., and / or other currently known or future communication protocols to be developed.

[0039] In the embodiments of the present application, the term "network device" refers to, for example, a device in a communication system that connects a terminal device to the communication network and provides services to the terminal device. Network devices may include, but are not limited to, the following devices: base station (BS), access point (AP), transmission reception point (TRP), broadcast transmitter, mobile management entity (MME), gateway, server, radio network controller (RNC), base station controller (BSC), etc.

[0040] Base stations may include, but are not limited to, NodeBs (NBs), evolved NodeBs (eNodeBs or eNBs), and 5G base stations (gNBs), among others. They may also include remote radio heads (RRHs), remote radio units (RRUs), relays, or low-power nodes (e.g., femto, pico, etc.). The term "base station" may include some or all of their functions, and each base station may provide communication coverage for a specific geographic area. The term "cell" may refer to a base station and / or its coverage area, depending on the context in which the term is used.

[0041] In the embodiments of the present application, the term "user equipment" (UE) refers to, for example, a device that accesses a communication network through a network device and receives network services, and may also be referred to as "terminal equipment" (TE). Terminal equipment may be fixed or mobile, and may also be referred to as a mobile station (MS), terminal, user, subscriber station (SS), access terminal (AT), station, etc.

[0042] Terminal devices may include, but are not limited to, the following devices: cellular phones, personal digital assistants (PDAs), wireless modems, wireless communication devices, handheld devices, machine-type communication devices, laptop computers, cordless phones, smartphones, smart watches, digital cameras, etc.

[0043] For another example, in scenarios such as the Internet of Things (IoT), the terminal device can also be a machine or device for monitoring or measurement, including but not limited to: machine type communication (MTC) terminal, vehicle-mounted communication terminal, device-to-device (D2D) terminal, machine-to-machine (M2M) terminal, and so on.

[0044] In the current channel state information generation and feedback process, the network device sends a channel state information reference signal (CSI-RS) to each terminal device. The terminal device estimates the channel using the received CSI-RS and obtains an estimate of the spatial channel matrix. The terminal device further uses the estimated spatial channel matrix to obtain CSI. In New Radio (NR), CSI feedback is implicit, meaning that the terminal device feeds back CSI in the form of recommended transmission parameters to the network device. These transmission parameters include channel state indicator (CQI), precoding matrix indicator (PMI), CSI-RS resource indicator (CRI), single-segment signaling (SSBRI), layer indicator (LI), rank indicator (RI), and physical layer RSRP (L1-RSRP). The network device can directly use the parameters recommended by the terminal device for downlink transmission, or it can choose not to use the recommended parameters.

[0045] Figure 1 is a schematic diagram of CSI generation and feedback based on the traditional codebook method. As shown in Figure 1, the network device sends a CSI-RS to the terminal device. The terminal device performs channel estimation and singular value decomposition (SVD), generates CSI using the traditional codebook generation method, and feeds it back to the network device.

[0046] Among them, when using the traditional codebook method to feedback CSI, when the rank of the spatial channel matrix estimated by the terminal device is greater than 1, the RI (if reported) fed back by the terminal device to the network device may be greater than 1. In this case, the PMI is a multi-rank codebook. In NR Rel-15, two codebooks, type I and type II, are defined. The former is a conventional precision codebook and can be used for SU-MIMO (single-user multiple input multiple output) and MU-MIMO (multi-user multiple input multiple output) transmission. The latter is a high-precision codebook, mainly used in MU-MIMO scenarios. The latter has higher accuracy than the former, but has higher overhead. Both NR codebooks use a parameterized codebook structure and are divided into two levels (W=W1W2), where W1 describes the long-term, wideband characteristics of the channel and contains an oversampled DFT beam (group); W2 describes the short-term, sub-band characteristics of the channel. For the above two codebooks, the selection method of W1 is the same. Regarding the selection of W2, the type I codebook consists of a weighted column selection vector, which selects a beam for the subband from the oversampled beams in W1. In the high-precision type II codebook, W2 is used to linearly combine the DFT beams in W1.

[0047] To address the excessive overhead of the Type II codebook, Rel-16 defines an enhanced Type II codebook (e-type II codebook). The e-type II codebook still uses a two-level structure: reporting a set of wideband beams and then adding a set of narrowband combining coefficients to each beam. The Rel-16 e-type II codebook is enhanced by leveraging frequency domain correlation to reduce reporting overhead. Furthermore, the e-type II CSI allows for a two-fold increase in the frequency domain granularity of PMI reporting.

[0048] At the 3GPP RAN 94 meeting, AI / ML for NR air interface was approved as a new study item (SI). Using AI / ML to enhance CSI feedback is a key use case. Unlike traditional codebook-based CSI feedback, AI / ML does not use a codebook to quantize spatial channel information. Instead, it uses AI / ML to process it and generate CSI.

[0049] Figure 2 is a schematic diagram of CSI generation and feedback based on AI / ML methods. As shown in Figure 2, unlike the traditional codebook-based CSI generation and feedback method in Figure 1, in the method of Figure 2, the terminal device uses an AI / ML model to generate CSI and feed it back to the network device.

[0050] At the 3GPP RAN 109-e meeting, the following consensus was reached on the structure of AI / ML models:

[0051]

[0052] FIG3 is a schematic diagram of the above-mentioned two-sided AI / ML model. As shown in FIG3 , the UE uses the AI / ML-based CSI generation portion of the model to process the input spatial channel to obtain CSI, and the network device uses the AI / ML-based CSI reconstruction portion of the model to reconstruct the obtained CSI to obtain the restored spatial channel as output. The bilateral AI / ML model shown in FIG3 is only an example, and the present application is not limited thereto. The AI / ML model can also be set only on the UE side or only on the network device side. For details, please refer to the relevant technology.

[0053] Various embodiments of the present application are described below with reference to the accompanying drawings. These embodiments are merely exemplary and are not intended to limit the present application.

[0054] Embodiments of the first aspect

[0055] The present invention provides a method for feedback of channel state information, which is described from the perspective of a terminal device. FIG4 is a schematic diagram of the method for feedback of channel state information according to an embodiment of the present invention. Referring to FIG4 , the method includes:

[0056] 401: A terminal device receives first information sent by a network device, where the first information instructs the terminal device to generate and feed back channel state information.

[0057] 402: The terminal device generates N layers of channel state information, where N ≥ 2, wherein channel state information of at least two layers of the N layers of channel state information is indicated by unequal numbers of bits.

[0058] 403: The terminal device sends the channel state information of the N layer to the network device.

[0059] It is worth noting that FIG4 above only schematically illustrates an embodiment of the present application, and the present application is not limited thereto. For example, the execution order of the various operations may be appropriately adjusted, and other operations may be added or some operations may be reduced. Those skilled in the art may make appropriate modifications based on the above content, and are not limited to the description of FIG4 above.

[0060] According to the above method of the embodiment of the present application, unequal numbers of feedback bits are allocated to two or more sub-channels, that is, different feedback accuracies are used for two or more sub-channels (or the feedback accuracies of two or more sub-channels are further different), thereby improving the gain of the system.

[0061] In an embodiment of the present application, the network device instructs the terminal device to generate and feedback CSI through the first information. The present application does not limit the implementation method of the first information. For example, the first information can be included in the RRC signaling sent by the network device to the terminal device. The first information can be reflected as settings related to the CSI reporting settings, etc. For details, please refer to the relevant technology.

[0062] In an embodiment of the present application, the channel state information of the above-mentioned N layers can be generated based on a codebook, for example, by the method shown in Figure 1, or can be generated based on an AI / ML model, for example, by the method shown in Figure 2.

[0063] The method of the embodiment of the present application is described below using an example in which a terminal device generates the above-mentioned N-layer channel state information based on an AI / ML model.

[0064] In an embodiment of the present application, the terminal device may generate the channel state information of the N layer according to a predefined or preconfigured rule or a network device configured rule (referred to as a first rule). The first rule includes but is not limited to one or a combination of the following:

[0065] Available AI / ML models;

[0066] Criteria and / or methods for selecting AI / ML models;

[0067] Criteria and / or methods for truncating CSI;

[0068] Approximate method for the number of feedback bits;

[0069] The maximum number of downlink transmission layers;

[0070] The maximum total number of feedback bits; and

[0071] The feedback order of CSI.

[0072] In the above and following descriptions of the embodiments of the present application, unless otherwise specified, "predefined" means specified in standard documents, but the present application is not limited to this. "Predefined" may also refer to the factory settings of the product; in addition, "preconfigured" means that the network device is pre-configured, which is static or semi-static, and the present application does not limit the configuration method; in addition, "network device configured" means that the network device is dynamically configured, for example, by sending control information or configuration information or indication information or system information, and the present application does not limit the configuration method.

[0073] In some embodiments, the network device and the terminal device reach the above-mentioned consensus through predefinition or preconfiguration or network device configuration. The terminal device generates the above-mentioned N-layer channel state information based on the above-mentioned consensus and feeds it back to the network device. The network device obtains the above-mentioned N-layer channel state information based on the above-mentioned consensus and sends data to the terminal device based on the recovered channel state information.

[0074] In the first rule described above, the number of available AI / ML models can be one or more. For example, the network device configures an AI / ML model for the terminal device and informs the terminal device that it can only use the configured AI / ML model; for another example, the network device configures multiple available models for the terminal device to choose from; for another example, the network device and the terminal device reach a consensus in advance and agree on the available AI / ML models, and the terminal device uses the agreed AI / ML model (the agreed AI / ML model is one), or the terminal device selects an AI / ML model from the agreed AI / ML models (the agreed AI / ML models are more than one).

[0075] In the first rule above, the criteria and / or method for truncating CSI include:

[0076] If the length of the CSI bit sequence is greater than the specified bit sequence length, the CSI bit sequence is truncated;

[0077] If the length of the CSI bit sequence is the same as the length of the prescribed bit sequence, the CSI bit sequence is retained.

[0078] The truncation operation on the CSI bit sequence may be performed according to a second rule that is predefined, preconfigured, or configured by the network device. The second rule may be:

[0079] The component at the first position in the CSI bit sequence is removed so that the length of the resulting CSI bit sequence is the same as the length of the specified bit sequence. The first position is one or more bits at any position in the CSI bit sequence. For example, the first position is the first two bits or the last two bits of the CSI bit sequence, etc. In addition, the specified bit sequence length may be specified in a standard document, preconfigured, predefined, or configured by a network device.

[0080] The above is an example of the second rule, and the present application is not limited thereto. For example, the second rule may also be:

[0081] An IDFT (inverse discrete Fourier transform) operation is performed on the CSI bit sequence, the component at the second position in the CSI bit sequence is removed, and a DFT (discrete Fourier transform) operation is performed on the resulting bit sequence, so that the length of the resulting CSI bit sequence is the same as the length of a specified bit sequence, where the second position is one or more digits at any position in the CSI bit sequence. For example, the second position is the first two digits or the last two digits of the CSI bit sequence, etc. In addition, the specified bit sequence length can be specified in a standard document, preconfigured, predefined, or configured by a network device.

[0082] In the first rule described above, the order of CSI feedback can be from largest to smallest CSI values, or from smallest to largest CSI values, or another order agreed upon by the network device and the terminal device, as long as the network device and the terminal device reach a consensus on this. The CSI values ​​can be eigenvalues ​​or singular values ​​of the spatial channel matrix, but this application is not limited thereto. Depending on the CSI generation method, the CSI values ​​can also be other values.

[0083] In the above-mentioned first rule, the AI / ML model can be a bilateral AI / ML model or a unilateral AI / ML model, wherein the bilateral AL / ML (artificial intelligence / machine learning) model means that the AI / ML model is set on the network device side and the terminal device side, as shown in Figure 3, and has an AI / ML-based CSI generation part on the terminal device side, and has an AI / ML-based CSI reconstruction part on the network device side; in addition, the unilateral AL / ML model means that the AI / ML model is set on the network device side or the terminal device side, for example, has an AI / ML-based CSI generation part on the terminal device side, or has an AI / ML-based CSI reconstruction part on the network device side.

[0084] In some embodiments, the network device is not aware of the AI / ML model used by the terminal device. The terminal device may also send an indication message to the network device to inform the network device of the number, index, or identifier of the AI / ML model used to generate the channel state information, so that the network device can post-process the received channel state information accordingly and restore the post-processed channel state information.

[0085] The following describes the method of an embodiment of the present application for generating CSI based on an AI / ML model with reference to specific examples.

[0086] FIG5 is a schematic diagram of an example of a method according to an embodiment of the present application. As shown in FIG5 , the method includes:

[0087] 501: The network device informs the terminal device of the AI / ML model that can be used. The specific bit allocation refers to the standard document;

[0088] 502: The terminal device feeds back CSI according to the configuration of the network device.

[0089] In an embodiment of the present application, in operation 501, the network device sends a CSI-RS to the terminal device, and configures or instructs the terminal device through the first information to use unequal numbers of bits to feedback the CSI of at least two layers of N layers (N≥2) of CSI, and the number of feedback bits of the CSI of each layer is executed in accordance with the provisions of the standard text; in operation 502, the terminal device receives the CSI-RS, performs channel estimation, and generates and feeds back the CSI of the above N layers according to the configuration or instruction of the above first information.

[0090] In some embodiments, in operation 501, the network device may inform the terminal device of the available AI / ML model through the above-mentioned first information; in operation 502, the terminal device may use the existing method to perform channel estimation. For details, please refer to the relevant technology, and the description is omitted here.

[0091] In the above embodiment, in operation 502, the terminal device can use unequal numbers of bits to feed back to the network device for two or more layers of CSI based on the configuration or indication of the above first information and the consensus reached with the network device (the above first rule).

[0092] In the above embodiment, the pre-agreed items, i.e., the first rule (known to both the terminal device and the network device, including standard documents), include: the AI / ML model number; the criteria for selecting the AI / ML model; the order of CSI vector feedback; and how the network device reads CSI. The specific steps are as follows:

[0093] S1: The network device configures four available AI / ML models for the terminal device, sorts the output bits of these four AI / ML models in descending order, and numbers them 1, 2, 3, and 4;

[0094] S2: The network device sends CSI-RS;

[0095] S3: The terminal device receives CSI-RS and measures the spatial channel. The rank of the spatial channel matrix is ​​calculated to be 2. The singular value σ of the spatial channel matrix is ​​calculated. k , k=1,2, and σ1>σ2>0;

[0096] S4: For the singular vectors corresponding to the two singular values, the terminal device selects the two models with the largest number of feedback bits among the four AI / ML models, namely Model 1 and Model 2, in accordance with the standard, assigns Model 1 to the singular vector corresponding to σ2, and assigns Model 2 to the singular vector corresponding to σ1.

[0097] S5: The terminal device compresses the two singular vectors using the determined AI / ML model.

[0098] S6: The terminal device feeds back the two compressed vectors to the network device in the order of σ1>σ2;

[0099] S7: The network device selects the appropriate reconstructed portion of the AI / ML model for decompression based on the received compressed vector and generates a precoding matrix according to the provisions of the standard text;

[0100] S8: The network device uses the above precoding matrix to send data to the terminal device.

[0101] In the above S7, according to the aforementioned first rule, the network device knows which two AI / ML models are used by the terminal device and the number of bits output by each AI / ML model.

[0102] In other embodiments, in operation 501, the network device informs the terminal device that only a pre-agreed AI / ML model can be used. The number of bits output by this AI / ML model is generally not less than the number of bits corresponding to the layer with the maximum number of feedback bits specified in the standard document. In this case, since the same AI / ML model is used for different vectors of the spatial channel matrix, their outputs also have the same number of bits. Furthermore, the AI / ML output is truncated according to the number of feedback bits for each vector specified in the standard document.

[0103] In the above embodiment, in operation 502, the terminal device receives the CSI-RS, performs channel estimation using existing methods, and generates and feeds back CSI in accordance with the configuration information of the network device, for example, in accordance with the instructions of the base station and the provisions of the standard text. For two or more layers of CSI, the terminal device can use different processing methods.

[0104] In the above embodiment, the pre-agreed items, i.e., the first rule (known to both the terminal device and the network device, including standard documents), include: AI / ML model; CSI vector truncation criteria and methods; CSI vector feedback order; how the network device reads CSI, and the method for lengthening truncated CSI vectors. The specific steps are as follows:

[0105] S1: The network device configures an available AI / ML model for the terminal device, such as AI / ML model A.

[0106] S2: The network device sends CSI-RS;

[0107] S3: The terminal device receives CSI-RS and measures the spatial channel. The rank of the spatial channel matrix is ​​calculated to be 3. The singular value σ of the spatial channel matrix is ​​calculated. k , k=1,2,3, and σ1>σ2>σ3>0, as well as the singular vectors corresponding to these three singular values;

[0108] S4: The terminal device uses the AI / ML model A configured on the network device to process the three singular vectors, for example, by compressing them.

[0109] S5: The terminal device reprocesses the processed results according to the provisions of the standard document;

[0110] For example, if the length of the bit sequence of the processed result is longer than the length of the bit sequence specified by the standard, the bit sequence of the processed result is truncated. The truncation method includes but is not limited to:

[0111] Delete the corresponding last bit; or

[0112] Perform IDFT on the desired truncated bit sequence, delete the components at pre-specified positions, and perform DFT on the resulting sequence.

[0113] For another example, when the length of the bit sequence of the processing result is the same as the length of the bit sequence specified by the standard, no operation is performed.

[0114] S6: The terminal device feeds back the three reprocessed vectors to the network device in the order of σ1>σ2>σ3;

[0115] S7: The network device calculates the original length of each singular vector based on the provisions of the standard text and the number of bits output by the AI / ML model that the configured terminal device should use;

[0116] S8: The network device performs data post-processing based on the received reprocessed vector and the original length of the singular vector calculated in the previous step;

[0117] For example, if the length of the bit sequence of the result obtained by the terminal device in S4 is longer than the length of the bit sequence specified by the standard, since the terminal device has already truncated the result in S5, the network device can perform a corresponding lengthening operation, for example:

[0118] Fill the corresponding last bit with 0; or

[0119] Perform IDFT on the bit sequence that needs to be lengthened, add 0s at pre-specified positions, and then perform DFT on the resulting sequence.

[0120] For another example, when the length of the bit sequence obtained by the terminal device in the above S4 is the same as the length of the bit sequence specified by the standard, since the terminal device does not perform a truncation operation, the network device does not perform any operation accordingly.

[0121] S9: The network device uses the reconstruction part of the AI / ML model to recover the spatial channel based on the post-processing results of the previous step and generate a precoding matrix.

[0122] S10: The network device uses the above precoding matrix to send data to the terminal device.

[0123] In the above S7, according to the aforementioned first rule, the network device knows which AI / ML model the terminal device uses and the number of bits output by the AI / ML model.

[0124] FIG6 is a schematic diagram of another example of a method according to an embodiment of the present application. As shown in FIG6 , the method includes:

[0125] 601: The network device switches to the CSI unequal bit feedback mode. The bit number allocation is determined by the terminal device and reported to the network device.

[0126] 602: The terminal device feeds back CSI according to the configuration of the network device, which includes a bit allocation method.

[0127] In an embodiment of the present application, in operation 601, the network device sends a CSI-RS to the terminal device, and configures or instructs the terminal device through the first information to use unequal numbers of bits to feedback the CSI of at least two layers of N layers (N≥2) of CSI, and the number of feedback bits of each layer of CSI is determined by the terminal device and reported to the network device; wherein the total number of feedback bits can be configured by the network device, or can also be specified in the standard; in operation 602, the terminal device receives the CSI-RS, performs channel estimation, and generates and feeds back the above-mentioned N layers of CSI according to the configuration or instruction of the above-mentioned first information.

[0128] In the above embodiment, in operation 602, the terminal device selects one of several possible unequal feedback bit numbers agreed in advance according to the instructions of the network device, and then uses unequal numbers of bits to feed back at least two layers of CSI to the network device, and feeds back the method of unequal feedback bit numbers used to the network device.

[0129] In the above embodiment, the pre-agreed items, i.e., the first rule (known to both the terminal device and the base station, including standard documents), include: the approximate method for the number of feedback bits; the number of AI / ML models; the criteria for selecting AI / ML models; the order of CSI vector feedback; how the network device reads CSI; and the number of bits output by each AI / ML model. The specific steps are as follows:

[0130] S1: The network device sends CSI-RS;

[0131] The network device may configure and notify the CSI transmission layer as 2, and the total number of feedback bits is 90 bits.

[0132] S2: The terminal receives CSI-RS and measures the spatial channel. The rank of the spatial channel matrix is ​​calculated to be 3. The singular value σ of the channel matrix is ​​calculated. k , k=1,2,3, and σ1>σ2>σ3>0,

[0133] Among them, according to the configuration of the network device, the terminal device selects the two largest singular values ​​and their corresponding singular vectors (σ1>σ2);

[0134] S3: For the singular vectors corresponding to the above two singular values, the terminal device generates the corresponding singular vectors according to the total number of feedback bits and the unequal ratio. The number of allocated feedback bits is 1.2379 ÷ 2.2379 × 90 = 49.78 and 1 ÷ 2.2379 × 90 = 40.21.

[0135] S4: The terminal device approximates (quantizes) the number of feedback bits, and the results are 50 bits and 40 bits respectively;

[0136] S5: The terminal device compresses the two singular vectors using a determined or selected AI / ML model based on the approximate number of feedback bits.

[0137] Each pair of AI / ML models corresponds to a feedback bit number. Assume that the network device and the terminal device agree on a total of eight pairs of AI / ML models, with feedback bit numbers of 24 bits, 30 bits, 36 bits, 42 bits, 48 ​​bits, 54 bits, 60 bits, and 660 bits, respectively. The corresponding AI / ML model indexes, identifiers, or numbers are 000, 001, 010, 011, 100, 101, 110, and 111, respectively.

[0138] The criterion for the terminal device to determine or select an AI / ML model may be: the approximation of the number of feedback bits calculated by the terminal device to the AI / ML model that is closest to the existing AI / ML model. In the above embodiment, the layer corresponding to σ1 is expected to feedback 50 bits and uses an AI / ML model with a 48-bit feedback (i.e., the AI / ML model with an index of 100). The layer corresponding to σ2 is expected to feedback 40 bits and uses an AI / ML model with a 42-bit feedback (i.e., the AI / ML model with an index of 011).

[0139] S6: The terminal device feeds back the two compressed vectors and the index of the AI / ML model to the network device in the order of σ1>σ2, which are 100 and 011 respectively.

[0140] S7: The network device decompresses the received compressed vector based on the index of the AI / ML model and generates a precoding matrix.

[0141] S8: The network device uses the above precoding matrix to send data to the terminal device.

[0142] In the above S7, the network device knows which AI / ML model the terminal device uses and the number of bits output by the AI / ML model based on the index of the AI / ML model fed back by the terminal device. In addition, the feedback order of CSI (compressed vector) may be specified by the standard, that is, the network device may also determine the feedback order of the compressed vector (CSI) fed back by the terminal device according to the provisions of the standard, and then receive the corresponding CSI. The present application is not limited to this, and the feedback order of CSI may also be agreed upon by the network device and the terminal device, and the present application does not limit the agreed method.

[0143] It is worth noting that the above example only uses the example of generating N layers (N ≥ 2) of CSI based on the AI / ML model to illustrate the embodiment of the present application, but the present application is not limited to this. For example, the execution order between the various operations can be appropriately adjusted, and other operations can be added or some operations can be reduced. Those skilled in the art can make appropriate modifications based on the above content, and are not limited to the description in the above example.

[0144] The method of the embodiment of the present application is described below by taking an example in which a terminal device generates the above-mentioned N layers of channel state information based on a codebook.

[0145] In an embodiment of the present application, a terminal device can receive first configuration information sent by a network device, and the first configuration information is used to configure the terminal device with the value range of the broadband amplitude, the value range of the sub-band amplitude, and the value range of the phase combining coefficient corresponding to the above-mentioned at least two layers of channel state information, wherein at least one of the value range of the broadband amplitude, the value range of the sub-band amplitude, and the value range of the phase combining coefficient corresponding to the channel state information of the at least two layers is different.

[0146] For example, the wideband amplitudes corresponding to the channel state information of the at least two layers have different value ranges, and / or the subband amplitudes corresponding to the channel state information of the at least two layers have different value ranges, and / or the phase combining coefficients corresponding to the channel state information of the at least two layers have different value ranges. That is, the CSI of at least two layers among the N layers of CSI are indicated using unequal numbers of bits due to at least one of the wideband amplitude value ranges, subband amplitude value ranges, and phase combining coefficient value ranges corresponding to the channel state information of the at least two layers being different.

[0147] In the above embodiment, there is no limitation on the implementation method of the first configuration information. For example, the first configuration information may be included in the RRC signaling or high-layer signaling sent by the network device to the terminal device. For details, please refer to the relevant technology.

[0148] In the above embodiment, the CSI is generated based on a codebook. The CSI may include a PMI. The PMI may include wideband information or wideband information and subband information.

[0149] Among them, the broadband information includes: the beam offset of the two-dimensional oversampled DFT beam; the beam selected from a group of beams determined according to the beam offset of the two-dimensional oversampled DFT beam; the index corresponding to the maximum amplitude beam; and the broadband amplitude; the subband information includes: the phase combining coefficient and the subband amplitude.

[0150] In the above embodiment, as described above, at least one of the wideband amplitude value range, subband amplitude value range, and phase combining coefficient value range corresponding to the CSI of at least two layers is different. That is, the wideband amplitude value range corresponding to the CSI of at least two layers is the same or different; the phase combining coefficient value range corresponding to the CSI of at least two layers is the same or different; and the subband amplitude value range corresponding to the CSI of at least two layers is the same or different.

[0151] The method of the embodiment of the present application for generating CSI based on a codebook is described below with reference to specific examples.

[0152] In some embodiments, a type II codebook is considered, with two transmission layers. According to the method of the embodiment of the present application, the following codebook design can be adopted.

[0153] In this embodiment, assuming the RI is 2 and the singular values ​​(eigenvalues) of the spatial channel matrix satisfy σ1 > σ2, the right singular vectors (eigenvectors) corresponding to these two singular values ​​are v1 and v2, respectively. In this embodiment, the layer with the larger singular value uses fewer bits for feedback than the other layer.

[0154] In this embodiment, the network device can configure the following parameters for the terminal device:

[0155] 1. Parameters N1 and N2:

[0156] The above parameters are configured by the upper layer. The optional configuration and the selection of the corresponding oversampling coefficients O1 and O2 are shown in Table 5.2.2.2.1-2. CSI-RS ) is 2N1N2.

[0157] Table 5.2.2.2.1-2: Supported configurations for (N1, N2) and (O1, O2)

[0158]

[0159] 2. Parameter L:

[0160] The above parameters are configured by the upper layer. CSI-RS =4, L = 2; when P CSI-RS >4, L∈{2,3,4}.

[0161] 3. Parameter N PSK :

[0162] The above parameters are configured by the upper layer, and the value range may be different for different transport layers, denoted as N PSK,l :l=1,2,…v, where v is the value of RI. An example is: N PSK,1 ∈{4,8},N PSK,2 ∈{4,16}.

[0163] 4. Whether to use sub-band amplitude:

[0164] The above parameters are configured by the upper layer and their values ​​can be true or false.

[0165] 5. RI reported by the terminal device:

[0166] The above parameters do not exceed 2. Let v be the RI reported by the terminal device, and each PMI value corresponds to the index i1 and i2 of the codebook, where

[0167]

[0168]

[0169] In the above embodiment, the case where v=2 and sub-band amplitude (true) is used is considered. According to the existing standard definition, both transmission layers use broadband reporting and sub-band reporting. The parameters that need to be reported are:

[0170] i1=[i 1,1 i 2,2 i 1,3,1 i 1,4,1 i 1,3,2 i 1,4,2 ]

[0171] i2=[i 2,1,1 i 2,2,1 i 2,1,2 i 2,2,2 ]

[0172] In the above embodiment, the terminal device reports the contents of i1 and i2 according to the existing standard process. The physical meaning of each parameter is described as follows:

[0173] 1)i 1,1 : Broadband reporting, for the beam offset of the two-dimensional oversampled DFT beam, there are a total of O1O2 possibilities, so it is necessary bits to report.

[0174] 2)i 1,2 : Broadband reporting, the beam offset i selected in the previous step 1,1 , then a wave array consisting of N1N2 beams is fixed. 1,2 Indicates that L beams are selected from these N1N2 beams, a total of Possible, so need bits to report.

[0175] 3)i 1,3,l : Broadband reporting, indicating the index corresponding to the maximum amplitude beam of layer l. There are 2L possibilities, so it is necessary bits to report.

[0176] 4)i 1,4,l : Broadband reporting, indicating the broadband range.

[0177] Unlike existing standards, which specify a total of eight possible wideband amplitudes, in this embodiment, the number of bits required for the two layers of wideband amplitudes may differ. The possible wideband amplitudes are listed in Tables 5.2.2.2.3-2a and 5.2.2.2.3-2b below. (Excluding the maximum amplitude beam), there are a total of 2L-1 wideband amplitudes that require feedback. Therefore, the first layer requires (2L-1)log24 = 2(2L-1) bits for reporting, and the second layer requires (2L-1)log216 = 4(2L-1) bits for reporting.

[0178] Table 5.2.2.2.3-2a:i 1,4,l A mapping of elements: arrive

[0179]

[0180] Table 5.2.2.2.3-2b:i 1,4,l A mapping of elements: arrive

[0181]

[0182] 5)i 2,1,l ,l=1,2,…v(v=2): Sub-band reporting, indicating the phase combining coefficient.

[0183] Different from the existing standards, the phase combination coefficient of each layer in this embodiment is The size of the value range (alphabet) is not equal, that is, N PSK,1 ≠N PSK,2 In this embodiment, N PSK,1 ∈{4,8},N PSK,2 ∈{4,8,16}. Phase combining coefficient The values ​​of are as follows:

[0184]

[0185] Among them, each layer (the lth layer) requires min(M l ,K (2) )log2N PSK,l +2(M l -min(M l ,K (2) ))-log2N PSK,l bits to report, where M l is satisfied i 1,4,l The number of K (2)The values ​​of are shown in Table 5.2.2.2.3-4 below:

[0186] Table 5.2.2.2.3-4: Full-resolution subband coefficients when subbandAmplitude is set to "true"

[0187] LK (2) 243446

[0188] 6)i 2,2,l ,l=1,2,…v(v=2): Sub-band reporting, indicating the sub-band amplitude.

[0189] The existing standard stipulates that only min(M l ,K (2) )-1 strongest amplitude beam has sub-band amplitude reported. Different from 3GPP38.214, the value range of sub-band amplitude of each layer in this embodiment can be unequal. In this embodiment, the possible sub-band amplitudes are given in Table 5.2.2.2.3-3 and Table 5.2.2.2.3-3b below, requiring min(M1,K (2) )-1+2(min(M2,K (2) )-1) bits are reported.

[0190] Table 5.2.2.2.3-3:i 2,2,l A mapping of elements: arrive

[0191]

[0192] Table 5.2.2.2.3-3b:i 2,2,l A mapping of elements: arrive

[0193]

[0194]

[0195] In summary, the total number of feedback bits required in this embodiment is:

[0196]

[0197] For example, if (N1, N2) = (8, 1), then from Table 5.2.2.2.1-2, (O1, O2) = (4, 1). If L = 4, then from Table 5.2.2.2.3-4, K (2) =6. Select N PSK,1 =4, N PSK,2=8, M1=5, M2=5. According to the above formula, the total number of feedback bits is also 89 bits.

[0198] According to the method of this embodiment, compared with the existing method, the number of feedback bits of this embodiment is also 89 bits. However, compared with the existing technology, the method of this embodiment uses unequal numbers of bits to report the wideband amplitude, subband amplitude, and combined phase of the two transmission layers, achieving different feedback accuracy. That is, under the premise that the total number of feedback bits is equal, the embodiment of the present application uses unequal bit reporting for at least two layers, achieving the goal of using fewer bits to report for the layer with a larger singular value and more bits to report for the layer with a smaller singular value.

[0199] In some other embodiments, the type II codebook is still considered, and the difference from the previous embodiment is that the subband reporting is false.

[0200] In this embodiment, assuming the RI is 2 and the singular values ​​(eigenvalues) of the spatial channel matrix satisfy σ1 > σ2, the right singular vectors (eigenvectors) corresponding to these two singular values ​​are v1 and v2, respectively. In this example, the layer with the larger singular value uses fewer bits for feedback than the other layer.

[0201] In this embodiment, the network device can configure the following parameters for the terminal device:

[0202] 1. Parameters N1 and N2:

[0203] The above parameters are configured by the upper layer. The optional configuration and the selection of the corresponding oversampling coefficients O1 and O2 are shown in Table 5.2.2.2.1-2. CSI-RS ) is 2N1N2.

[0204] 2. Parameter L:

[0205] The above parameters are configured by the upper layer. CSI-RS =4, L = 2; when P CSI-RS >4, L∈{2,3,4}.

[0206] 3. Parameter N PSK :

[0207] The above parameters are configured by the upper layer.

[0208] 4. Whether to use sub-band amplitude:

[0209] The above parameters are configured by the upper layer and their values ​​can be true or false.

[0210] 5. RI reported by the terminal device:

[0211] The above parameters do not exceed 2. Let v be the RI reported by the terminal device, and each PMI value corresponds to the index i1 and i2 of the codebook, where

[0212]

[0213]

[0214] In the above embodiment, the case where v=2 and sub-band amplitude (false) is used is considered. According to the existing standard definition, both transmission layers use sub-band amplitude. The parameters that need to be reported are:

[0215] i1=[i 1,1 i 1,2 i 1,3,1 i 1,4,1 i 1,3,2 i 1,4,2 ]

[0216] In the above embodiment, the content reported by the terminal device to i1 is given by the existing standard process. The physical meaning of each parameter is described as follows:

[0217] 1)i 1,1 : Broadband reporting, for the beam offset of the two-dimensional oversampled DFT beam, there are a total of O1O2 possibilities, so it is necessary bits to report.

[0218] 2)i 1,2 : Broadband reporting, the beam offset i selected in the previous step 1,1 , then a wave array consisting of N1N2 beams is fixed. 1,2 Indicates that L beams are selected from these N1N2 beams, a total of Possible, so need bits to report.

[0219] 3)i 1,3,l : Broadband reporting, indicating the index corresponding to the maximum amplitude beam of layer l. There are 2L possibilities, so it is necessary bits to report.

[0220] 4)i 1,4,l : Broadband reporting, indicating the broadband range.

[0221] Unlike existing standards, which specify a total of eight possible wideband amplitudes, in this embodiment, the number of bits required for the wideband amplitudes of at least two layers can be unequal. The possible wideband amplitudes are listed in Tables 5.2.2.2.3-2a and 5.2.2.2.3-2b. (Excluding the maximum amplitude beam), there are a total of 2L-1 wideband amplitudes that require feedback. Therefore, the first layer requires (2L-1)log24 = 2(2L-1) bits for reporting, and the second layer requires (2L-1)log216 = 4(2L-1) bits for reporting.

[0222] In summary, the total number of feedback bits required in this embodiment is:

[0223]

[0224] For example, if (N1, N2) = (8, 1), then from Table 5.2.2.2.1-2, (O1, O2) = (4, 1). If L = 4, then according to the above formula, the total number of feedback bits is 57. In this embodiment, although the total number of feedback bits is equal to the total number of feedback bits calculated according to existing standards, the broadband reporting accuracy of the two transmission layers is different, and the two layers feedback unequal numbers of bits.

[0225] It is worth noting that the above example only uses the example of generating N layers (N ≥ 2) of CSI based on the codebook to illustrate the embodiment of the present application, but the present application is not limited to this. The network device can also configure other parameters for the terminal device or reduce the configuration of certain parameters. Those skilled in the art can make appropriate modifications based on the above content and are not limited to the description in the above example.

[0226] In an embodiment of the present application, the network device may further instruct the terminal device to report CSI feedback parameters according to the reportQuatity (report quantity) specified by the network device.

[0227] In the above embodiment, the terminal device receives the second configuration information sent by the network device, and the second configuration information includes: reportQuatity that needs to be fed back, and the reportQuatity that needs to be fed back includes at least one of the following: the number of feedback bits of the above-mentioned at least two layers of channel state information, and the index, number or identifier of the AI / ML model used by each of the above-mentioned at least two layers of channel state information.

[0228] FIG7 is a schematic diagram of another example of a method according to an embodiment of the present application. As shown in FIG7 , the method includes:

[0229] 701: The network device configures the reportQuantity required for feedback for the terminal device, indicating whether to use unequal bits to feedback CSI;

[0230] 702: The terminal device feeds back CSI according to the configuration of the network device.

[0231] In an embodiment of the present application, in operation 701, the network device sends a CSI-RS to the terminal device, and configures or instructs the terminal device through the first information to report the parameters of the channel state information feedback according to the reportQuantity specified by the network device; in operation 702, the terminal device receives the CSI-RS, performs channel estimation, and generates and feeds back the above-mentioned N-layer CSI according to the configuration or instruction of the above-mentioned first information.

[0232] In some embodiments, in operation 701, reportQuantity includes whether unequal numbers of bits are used to feedback CSI of two or more layers. If unequal numbers of bits are used for feedback, the specific number of bits for feedback of CSI of each layer is implemented in accordance with the provisions of the standard.

[0233] In the above embodiment, in operation 702, the terminal device decides to use equal or unequal number of bits to feedback CSI of different layers, and reports it to the network device.

[0234] It is worth noting that the above example only schematically illustrates the reportQuatity that requires feedback, but the present application is not limited thereto. The reportQuatity that requires feedback may also include other content. Those skilled in the art may make appropriate modifications based on the above content, and are not limited to the description in the above example.

[0235] The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

[0236] According to the method of the embodiment of the present application, unequal numbers of feedback bits are allocated to two or more sub-channels, that is, different feedback precisions are used for the two or more sub-channels, thereby improving system gain. In addition, compared with the traditional codebook method, the feedback precision is further improved.

[0237] Embodiments of the second aspect

[0238] The embodiment of the present application provides a data transmission method, which is described from the perspective of a network device. This method is a network-side process corresponding to the method of the embodiment of the first aspect, wherein the same contents as those of the embodiment of the first aspect are not repeated.

[0239] FIG8 is a schematic diagram of a data sending method according to an embodiment of the present application. As shown in FIG8 , the method includes:

[0240] 801: The network device sends first information to the terminal device, where the first information instructs the terminal device to generate and feed back channel state information.

[0241] 802: The network device receives N layers of channel state information sent by the terminal device, where N is greater than or equal to 2, and channel state information of at least two layers of the N layers of channel state information is indicated by unequal numbers of bits.

[0242] 803: The network device sends data to the terminal device according to the channel state information of the N layer.

[0243] In operation 803, the network device may generate a precoding matrix based on the channel state information of the N layers, apply the precoding matrix to the data to be transmitted by the network device to the terminal device, and transmit the data to the terminal device after the precoding matrix is ​​applied. The specific transmission process can be referred to in related technologies and is not described here.

[0244] In an embodiment of the present application, the channel state information of the above-mentioned N layers can be generated by the terminal device based on a codebook, or can be generated by the terminal device based on an AL / ML model.

[0245] The method of the embodiment of the present application is described below using an example in which a terminal device generates N layers (N≥2) of channel state information based on an AI / ML model.

[0246] In some embodiments, the first information includes a first rule, that is, the network device configures the first rule for the terminal device, wherein the first rule is a rule of using unequal numbers of bits to indicate the channel state information of the at least two layers. Alternatively, the first rule is predefined or preconfigured.

[0247] In the above embodiment, the first rule may include but is not limited to one or a combination of the following:

[0248] Available AI / ML models;

[0249] Criteria and / or methods for selecting AI / ML models;

[0250] Criteria and / or methods for truncating CSI;

[0251] Approximate method for the number of feedback bits;

[0252] The maximum number of downlink transmission layers;

[0253] Maximum total number of feedback bits;

[0254] The feedback order of CSI.

[0255] In some embodiments, the number of AI / ML models available in the first rule is one or more.

[0256] In some embodiments, the criteria and / or method for truncating CSI in the first rule include:

[0257] If the length of the CSI bit sequence is greater than the specified bit sequence length, the CSI bit sequence is truncated;

[0258] If the length of the CSI bit sequence is the same as the length of the prescribed bit sequence, the CSI bit sequence is retained.

[0259] The CSI bit sequence may be truncated according to a second rule that is predefined, preconfigured, or configured by a network device. The second rule includes but is not limited to:

[0260] Removing the component at the first position in the CSI bit sequence so that the length of the obtained CSI bit sequence is the same as the length of the specified bit sequence, wherein the first position is one or more bits at any position in the CSI bit sequence, or,

[0261] An IDFT operation is performed on the CSI bit sequence, the component at the second position in the CSI bit sequence is removed, and a DFT operation is performed on the resulting bit sequence so that the length of the resulting CSI bit sequence is the same as the length of a specified bit sequence, where the second position is one or more bits at any position in the CSI bit sequence.

[0262] In some embodiments, the order of CSI feedback in the first rule includes: an order from large to small CSI values; or an order from small to large CSI values; or another order agreed upon by the network device and the terminal device. The CSI values ​​may be eigenvalues ​​or singular values ​​of the spatial channel matrix.

[0263] In some embodiments, the AI / ML model is a bilateral AI / ML model or a unilateral AI / ML model. The bilateral AL / ML model means that the AI / ML model is set on the network device side and the terminal device side; the unilateral AL / ML model means that the AI / ML model is set on the network device side or the terminal device side.

[0264] In some embodiments, if there is more than one AI / ML model available, the network device uses the corresponding AI / ML model to recover the received channel state information based on the available AI / ML models and the criteria and / or method for selecting the AI / ML model.

[0265] In some embodiments, there is only one available AI / ML model, and the network device post-processes the received channel state information based on the first rule; and restores the post-processed channel state information.

[0266] In some embodiments, the network device may also receive indication information from the terminal device, which indicates the number, index, or identifier of the AI / ML model used for the above-mentioned channel state information; the network device may post-process the received channel state information according to the number, index, or identifier of the AI / ML model used for the channel state information; and restore the post-processed channel state information.

[0267] In the aforementioned embodiments, post-processing may include:

[0268] If the length of the bit sequence of the CSI processed by the terminal device is greater than the prescribed bit sequence length, since the terminal device has performed a truncation operation on the processed CSI, the network device may determine a corresponding padding method according to the criterion and / or method for truncation of the CSI included in the first rule, and perform the padding operation on the bit sequence of the received CSI;

[0269] If the length of the bit sequence of the CSI processed by the terminal device is the same as the length of the specified bit sequence, since the terminal device does not perform a truncation operation on the processed CSI, the network device does not perform a corresponding lengthening operation on the received CSI, but retains the bit sequence of the CSI.

[0270] In the above embodiment, performing the padding operation on the CSI bit sequence according to the first rule may include:

[0271] Fill the corresponding bit at the first position with 0 or 1 or a sequence generated according to a predefined rule; or

[0272] Perform IDFT operation on the bit sequence that needs to be padded, fill the corresponding bit position at the second position with 0 or 1 or a sequence generated according to predefined rules, and then perform DFT operation on the resulting sequence.

[0273] The method of the embodiment of the present application is described below by taking an example in which a terminal device generates N layers (N≥2) of channel state information based on a codebook.

[0274] The network device can also configure the terminal device with the value ranges of the broadband amplitude, the value ranges of the sub-band amplitude, and the value ranges of the phase combining coefficient corresponding to the above-mentioned at least two layers of channel state information, wherein at least one of the value ranges of the broadband amplitude, the value ranges of the sub-band amplitude, and the value ranges of the phase combining coefficient corresponding to the at least two layers of channel state information is different.

[0275] In some embodiments, the channel state information includes PMI, which includes broadband information (such as i1) or broadband information (such as i1) and subband information (such as i2). The broadband information (such as i1) includes: the beam offset of the two-dimensional oversampled DFT beam (such as i2); 1,1 ); A beam selected from a set of beams determined according to the beam offset of the two-dimensional oversampled DFT beam (such as i 1,2 ); the index corresponding to the maximum amplitude beam (such as i 1,3,l ); and broadband amplitude (such as i 1,4,l ), wherein the value ranges of the broadband amplitudes corresponding to the channel state information of the at least two layers are the same or different; in addition, the sub-band information (such as i2) includes: a phase combining coefficient (such as i 2,1,l ) and sub-band amplitude (such as i 2,2,l ), wherein the value ranges of the phase combining coefficients corresponding to the channel state information of the at least two layers are the same or different; the value ranges of the sub-band amplitudes corresponding to the channel state information of the at least two layers are the same or different.

[0276] The method of the embodiment of the present application is described above using the example of a terminal device generating CSI based on an AI / ML model and generating CSI based on a codebook. In other embodiments, the network device may further configure the terminal device with a reportQuatity that requires feedback. The reportQuatity includes at least one of the following: the number of feedback bits of the channel state information of at least two layers, and the index, number, or identifier of the AI / ML model used for each of the at least two layers of channel state information. Thus, the terminal device can provide corresponding feedback based on the configuration of the network device.

[0277] The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

[0278] The above only describes the steps or processes related to the present application, but the present application is not limited thereto. The data sending method may also include other steps or processes, and the specific contents of these steps or processes may be referred to the prior art.

[0279] According to the method of the embodiment of the present application, unequal numbers of feedback bits are allocated to two or more sub-channels, that is, different feedback precisions are used for the two or more sub-channels, thereby improving system gain. In addition, compared with the traditional codebook method, the feedback precision is further improved.

[0280] Embodiments of the third aspect

[0281] An embodiment of the present application provides a device for feedback of channel state information, which may be, for example, a terminal device, or may be one or more components or assemblies configured in the terminal device.

[0282] Figure 9 is a schematic diagram of a channel state information feedback apparatus according to an embodiment of the present application. Since the principle of solving the problem of this apparatus is similar to that of the method according to the embodiment of the first aspect, its specific implementation can refer to the implementation of the method according to the embodiment of the first aspect, and the same content will not be repeated here. As shown in Figure 9, the channel state information feedback apparatus 900 according to the embodiment of the present application includes: a receiving unit 901, a generating unit 902, and a sending unit 903.

[0283] The receiving unit 901 is used to receive first information sent by a network device, which first information instructs a terminal device to generate and feed back channel state information; the generating unit 902 is used to generate N layers of channel state information, N≥2, wherein the channel state information of at least two layers of the N layers of channel state information is indicated by unequal numbers of bits; the sending unit 903 is used to send the N layers of channel state information to the network device.

[0284] In some embodiments, the channel state information of the N layers is generated based on a codebook or an AI / ML model.

[0285] In some embodiments, the generating unit 902 generates the channel state information of the N layers according to a first rule that is predefined or preconfigured or configured by the network device, where the first rule includes one or a combination of the following:

[0286] Available AI / ML models;

[0287] Criteria and / or methods for selecting AI / ML models;

[0288] Criteria and / or methods for truncating CSI;

[0289] Approximate method for the number of feedback bits;

[0290] The maximum number of downlink transmission layers;

[0291] Maximum total number of feedback bits;

[0292] The feedback order of CSI.

[0293] In the above embodiments, the number of available AI / ML models is one or more.

[0294] In the above embodiment, the criteria and / or method for truncating CSI include:

[0295] If the length of the CSI bit sequence is greater than the specified bit sequence length, the CSI bit sequence is truncated;

[0296] If the length of the CSI bit sequence is the same as the length of the prescribed bit sequence, the CSI bit sequence is retained.

[0297] The truncating operation on the CSI bit sequence includes: truncating the CSI bit sequence according to a second rule that is predefined or preconfigured or configured by a network device, where the second rule includes:

[0298] Removing the component at the first position in the CSI bit sequence so that the length of the obtained CSI bit sequence is the same as the length of the specified bit sequence, wherein the first position is one or more bits at any position in the CSI bit sequence, or,

[0299] An IDFT operation is performed on the CSI bit sequence, the component at the second position in the CSI bit sequence is removed, and a DFT operation is performed on the resulting bit sequence so that the length of the resulting CSI bit sequence is the same as the length of a specified bit sequence, where the second position is one or more bits at any position in the CSI bit sequence.

[0300] In the above embodiment, the CSI feedback order includes: the order of CSI values ​​from largest to smallest; or the order of CSI values ​​from smallest to largest, or another order agreed upon by the network device and the terminal device. The CSI values ​​include: eigenvalues ​​or singular values ​​of the spatial channel matrix.

[0301] In some embodiments, the AI / ML model is a bilateral AI / ML model or a unilateral AI / ML model. The bilateral AL / ML model means that the AI / ML model is set on the network device side and the terminal device side; the unilateral AL / ML model means that the AI / ML model is set on the network device side or the terminal device side.

[0302] In some embodiments, the sending unit 903 also sends indication information to the network device, which indicates the number, index or identifier of the AI / ML model used for the above-mentioned channel state information, so that the network device post-processes the received channel state information according to the number, index or identifier of the AI / ML model used for the channel state information, and restores the post-processed channel state information.

[0303] In some embodiments, the receiving unit 901 also receives first configuration information sent by the network device, and the first configuration information is used to configure the value range of the broadband amplitude, the value range of the sub-band amplitude, and the value range of the phase combining coefficient corresponding to at least two layers of channel state information for the terminal device, wherein at least one of the value range of the broadband amplitude, the value range of the sub-band amplitude, and the value range of the phase combining coefficient corresponding to at least two layers of channel state information is different.

[0304] In the above embodiment, the channel state information includes PMI, which includes broadband information (i1) or broadband information (i1) and subband information (i2). The broadband information (i1) includes: the beam offset (i2) of the two-dimensional oversampled DFT beam; 1,1 ); A beam selected from a set of beams determined based on a beam offset of a two-dimensional oversampled DFT beam (i 1,2 ); the index corresponding to the maximum amplitude beam (i 1,3,l ); and broadband amplitude (i 1,4,l ), wherein the value ranges of the broadband amplitudes corresponding to the channel state information of at least two layers are the same or different. The subband information (i2) includes: the phase combining coefficient (i 2,1,l ) and sub-band amplitude (i 2,2,l ), wherein the value ranges of the phase combining coefficients corresponding to the channel state information of at least two layers are the same or different; and the value ranges of the sub-band amplitudes corresponding to the channel state information of at least two layers are the same or different.

[0305] In some embodiments, the receiving unit 901 also receives second configuration information sent by the network device, where the second configuration information includes: reportQuatity that requires feedback, where the reportQuatity that requires feedback includes at least one of the following: the number of feedback bits of at least two layers of channel state information, and the index, number, or identifier of the AI / ML model used by each of the at least two layers of channel state information.

[0306] It is worth noting that the above description only describes the components or modules related to the present application, but the present application is not limited thereto. The channel state information feedback apparatus 900 may also include other components or modules. For details of these components or modules, please refer to the relevant art.

[0307] For simplicity, FIG9 illustrates only the connection relationships and signal paths between various components or modules. However, those skilled in the art will appreciate that various related technologies, such as bus connections, may be employed. The aforementioned components or modules may be implemented using hardware such as processors and memory; this is not a limitation of the present invention.

[0308] The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

[0309] According to the embodiments of the present application, unequal numbers of feedback bits are allocated to two or more sub-channels, that is, different feedback accuracies are used for the two or more sub-channels, thereby improving system gain. In addition, compared with the traditional codebook method, the feedback accuracy is further improved.

[0310] Embodiments of the fourth aspect

[0311] An embodiment of the present application provides a data sending device, which may be, for example, a network device, or may be one or more components or assemblies configured on the network device.

[0312] FIG10 is a schematic diagram of a data transmission device according to this embodiment. Since the principle of solving the problem of this device is similar to that of the method according to the second aspect, its specific implementation can refer to the implementation of the method according to the second aspect, and the same contents will not be repeated here. As shown in FIG10 , the data transmission device 1000 according to this embodiment includes: a first transmitting unit 1001, a receiving unit 1002, and a second transmitting unit 1003.

[0313] Among them, the first sending unit 1001 is used to send first information to the terminal device, and the first information instructs the terminal device to generate and feedback channel state information; the receiving unit 1002 is used to receive N layers of channel state information sent by the terminal device, N≥2, and the channel state information of at least two layers of the N layers of channel state information is indicated by unequal numbers of bits; the second sending unit 1003 is used to send data to the terminal device according to the above-mentioned N layers of channel state information.

[0314] In some embodiments, the second sending unit 1003 sends data to the terminal device according to the channel state information of the N layer, including:

[0315] Generate a precoding matrix according to the channel state information of the N layers;

[0316] Applying the precoding matrix to data sent from a network device to a terminal device;

[0317] Send the data after the precoding matrix is ​​applied to the terminal device.

[0318] In some embodiments, the channel state information of the N layers is generated based on a codebook, or based on an AL / ML model.

[0319] In some embodiments, the first information includes a first rule for indicating channel state information of the at least two layers using unequal numbers of bits.

[0320] In some embodiments, the first rule is predefined or preconfigured.

[0321] In the above embodiment, the first rule includes one or a combination of the following:

[0322] Available AI / ML models;

[0323] Criteria and / or methods for selecting AI / ML models;

[0324] Criteria and / or methods for truncating CSI;

[0325] Approximate method for the number of feedback bits;

[0326] The maximum number of downlink transmission layers;

[0327] Maximum total number of feedback bits;

[0328] The feedback order of CSI.

[0329] In some embodiments, the number of AI / ML models available in the first rule is one or more.

[0330] In some embodiments, the criteria and / or method for truncating CSI in the first rule include:

[0331] If the length of the CSI bit sequence is greater than the specified bit sequence length, the CSI bit sequence is truncated;

[0332] If the length of the CSI bit sequence is the same as the length of the prescribed bit sequence, the CSI bit sequence is retained.

[0333] In the above embodiment, performing a truncation operation on the CSI bit sequence includes: performing a truncation operation on the CSI bit sequence according to a second rule that is predefined or preconfigured or configured by a network device, where the second rule includes:

[0334] Removing the component at the first position in the CSI bit sequence so that the length of the obtained CSI bit sequence is the same as the length of the specified bit sequence, wherein the first position is one or more bits at any position in the CSI bit sequence, or,

[0335] Perform an IDFT operation on the CSI bit sequence, remove the component at the second position in the CSI bit sequence, and perform a DFT operation on the resulting bit sequence so that the length of the resulting CSI bit sequence is the same as the length of a specified bit sequence, where the second position is one or more bits at any position in the CSI bit sequence.

[0336] In some embodiments, the CSI feedback order in the first rule includes:

[0337] The order of the CSI values ​​from largest to smallest; or

[0338] The values ​​corresponding to CSI are in ascending order.

[0339] In the above embodiment, the value corresponding to the CSI may include: an eigenvalue or a singular value of the spatial channel matrix.

[0340] In some embodiments, the AI / ML model is a bilateral AI / ML model or a unilateral AI / ML model. A bilateral AL / ML model means that the AI / ML model is set on the network device side and the terminal device side; a unilateral AL / ML model means that the AI / ML model is set on the network device side or the terminal device side.

[0341] In some embodiments, there are more than one AI / ML models available, and the second sending unit 1003 uses the corresponding AI / ML model to recover the received channel state information based on the available AI / ML models and the criteria and / or method for selecting the AI / ML model.

[0342] In some embodiments, there is only one available AI / ML model, and the second sending unit 1003 post-processes the received channel state information based on the first rule; and restores the post-processed channel state information.

[0343] In some embodiments, the receiving unit 1002 also receives indication information, which indicates the number, index, or identifier of the AI / ML model used by the channel state information; the second sending unit 1003 post-processes the received channel state information according to the number, index, or identifier of the AI / ML model used by the channel state information, and restores the post-processed channel state information.

[0344] In the aforementioned embodiments, the post-processing performed by the second sending unit 1003 includes:

[0345] If the length of the bit sequence of the CSI after processing by the terminal device is greater than the length of the prescribed bit sequence, then since the terminal device has performed a truncation operation on the CSI, the second sending unit 1003 may determine the corresponding lengthening method according to the criteria and / or method for truncating the CSI contained in the first rule, and perform a lengthening operation on the bit sequence of the received CSI; if the length of the bit sequence of the CSI after processing by the terminal device is the same as the length of the prescribed bit sequence, then since the terminal device has not performed a truncation operation on the CSI, the second sending unit 1003 retains the bit sequence of the CSI.

[0346] In the above embodiment, the second sending unit 1003 performs a padding operation on the bit sequence of the CSI according to the first rule, including: padding the corresponding bit position at the first position with 0 or 1 or a sequence generated according to a predefined rule; or performing an IDFT operation on the bit sequence to be padded, padding the corresponding bit position at the second position with 0 or 1 or a sequence generated according to a predefined rule, and then performing a DFT operation on the resulting sequence.

[0347] In some embodiments, as shown in FIG10 , the apparatus 1000 further includes:

[0348] The first configuration unit 1004 configures the terminal device with the value ranges of the broadband amplitude, the value ranges of the sub-band amplitude, and the value ranges of the phase combining coefficient corresponding to at least two layers of channel state information, wherein at least one of the value ranges of the broadband amplitude, the value ranges of the sub-band amplitude, and the value ranges of the phase combining coefficient corresponding to the at least two layers of channel state information is different.

[0349] In the above embodiment, the channel state information includes PMI, and the PMI includes broadband information (i1) or broadband information (i1) and subband information (i2); the broadband information (i1) includes: a beam offset (i2) of a two-dimensional oversampled DFT beam; 1,1 ); A beam selected from a set of beams determined according to the beam offset of the two-dimensional oversampled DFT beam (i 1,2 ); the index corresponding to the maximum amplitude beam (i 1,3,l ); and broadband amplitude (i 1,4,l ), wherein the value ranges of the broadband amplitudes corresponding to the channel state information of at least two layers are the same or different; the sub-band information (i2) includes: a phase combining coefficient (i 2,1,l ) and sub-band amplitude (i 2,2,l ), wherein the value ranges of the phase combining coefficients respectively corresponding to the channel state information of at least two layers are the same or different, and the value ranges of the sub-band amplitudes respectively corresponding to the channel state information of at least two layers are the same or different.

[0350] In some embodiments, as shown in FIG10 , the apparatus 1000 further includes:

[0351] The second configuration unit 1005 configures the terminal device with a reportQuatity that requires feedback, where the reportQuatity includes at least one of the following: the number of feedback bits of at least two layers of channel state information, and the index, number, or identifier of the AI / ML model used by each of the at least two layers of channel state information.

[0352] It is worth noting that the above description only describes the components or modules related to the present application, but the present application is not limited thereto. The data sending device 1000 may also include other components or modules. For the specific contents of these components or modules, reference may be made to the relevant art.

[0353] For simplicity, FIG10 illustrates only the connection relationships and signal paths between various components or modules. However, those skilled in the art will appreciate that various related technologies, such as bus connections, may be employed. The aforementioned components or modules may be implemented using hardware such as processors and memories; this is not a limitation of the present invention.

[0354] The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

[0355] According to the embodiments of the present application, unequal numbers of feedback bits are allocated to two or more sub-channels, that is, different feedback accuracies are used for the two or more sub-channels, thereby improving system gain. In addition, compared with the traditional codebook method, the feedback accuracy is further improved.

[0356] Embodiments of the fifth aspect

[0357] An embodiment of the present application provides a communication system.

[0358] Figure 11 is a schematic diagram of a communication system according to an embodiment of the present application. As shown in Figure 11, the communication system 1100 according to the embodiment of the present application includes a network device 1101 and a terminal device 1102. For simplicity, Figure 11 only uses one terminal device and one network device as an example, but the embodiments of the present application are not limited thereto.

[0359] In the embodiment of the present application, existing services or future services can be transmitted between the network device 1101 and the terminal device 1102. For example, these services may include, but are not limited to, enhanced mobile broadband (eMBB), massive machine type communication (mMTC), ultra-reliable low-latency communication (URLLC), and vehicle-to-everything (V2X) communication, etc.

[0360] In some embodiments, network device 1101 can be the network device described in the embodiment of the fourth aspect, and is configured to execute the method described in the embodiment of the second aspect. The specific content has been described in the embodiments of the second aspect and the fourth aspect, and its content is merged here and will not be repeated here.

[0361] In some embodiments, the terminal device 1102 can be the terminal device described in the embodiment of the third aspect, and is configured to execute the method described in the embodiment of the first aspect. The specific content has been described in the embodiment of the first aspect and the embodiment of the third aspect, and its content is merged here and will not be repeated here.

[0362] An embodiment of the present application further provides a terminal device, which may be, for example, a UE, but the present application is not limited thereto and may also be other devices.

[0363] Figure 12 is a schematic diagram of a terminal device according to an embodiment of the present application. As shown in Figure 12, terminal device 1200 according to an embodiment of the present application includes a processor (e.g., a central processing unit (CPU)) 1201 and a memory 1202. Memory 1202 stores data and programs and is coupled to processor 1201. It should be noted that this diagram is exemplary; other types of structures may be used to supplement or replace this structure to implement telecommunication or other functions.

[0364] For example, the processor 1201 may be configured to execute a program to implement the method described in the embodiment of the first aspect.

[0365] As shown in Figure 12 , the terminal device 1200 may further include: a communication module 1203, an input unit 1204, a display 1205, and a power supply 1206. The functions of these components are similar to those in the prior art and are not described in detail here. It is worth noting that the terminal device 1200 does not necessarily include all of the components shown in Figure 12 , and these components are not essential. Furthermore, the terminal device 1200 may also include components not shown in Figure 12 , for which reference may be made to related art.

[0366] An embodiment of the present application also provides a network device, which may be, for example, a base station (gNB), but the present application is not limited thereto and may also be other network devices.

[0367] Figure 13 is a schematic diagram of a network device according to an embodiment of the present application. As shown in Figure 13 , network device 1300 according to an embodiment of the present application includes a processor (e.g., a central processing unit (CPU)) 1301 and a memory 1302 ; memory 1302 is coupled to processor 1301 . Memory 1302 can store various data and also stores information processing programs, which are executed under the control of CPU 1301 .

[0368] For example, the processor 1301 may be configured to execute a program to implement the method as described in the embodiment of the second aspect.

[0369] In addition, as shown in FIG13 , network device 1300 may further include: a transceiver 1303 and an antenna 1304; wherein, the functions of the above components are similar to those in the prior art and are not described in detail here. It is worth noting that network device 1300 does not necessarily include all the components shown in FIG13 ; in addition, network device 1300 may also include components not shown in FIG13 , and reference may be made to the prior art for details.

[0370] An embodiment of the present application further provides a computer-readable program, wherein when the program is executed in a terminal device, the program enables a computer to execute the method described in the embodiment of the first aspect in the terminal device.

[0371] An embodiment of the present application further provides a storage medium storing a computer-readable program, wherein the computer-readable program enables a computer to execute the method described in the embodiment of the first aspect in a terminal device.

[0372] An embodiment of the present application further provides a computer-readable program, wherein when the program is executed in a network device, the program enables a computer to execute the method described in the embodiment of the second aspect in the network device.

[0373] An embodiment of the present application further provides a storage medium storing a computer-readable program, wherein the computer-readable program enables a computer to execute the method described in the embodiment of the second aspect in a network device.

[0374] The above devices and methods of the present application can be implemented by hardware or by a combination of hardware and software. The present application relates to such a computer-readable program that, when executed by a logic component, enables the logic component to implement the devices or components described above, or enables the logic component to implement the various methods or steps described above. The logic component is, for example, a field programmable logic component, a microprocessor, a processor used in a computer, etc. The present application also relates to a storage medium for storing the above program, such as a hard disk, a magnetic disk, an optical disk, a DVD, a flash memory, etc.

[0375] The method / device described in conjunction with the embodiments of the present application can be directly embodied as hardware, a software module executed by a processor, or a combination of the two. For example, one or more of the functional block diagrams shown in the figure and / or one or more combinations of functional block diagrams can correspond to various software modules of the computer program flow or to various hardware modules. These software modules can respectively correspond to the various steps shown in the figure. These hardware modules can be implemented by solidifying these software modules, for example, using a field programmable gate array (FPGA).

[0376] The software module may be located in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. A storage medium may be coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium; or the storage medium may be an integral part of the processor. The processor and the storage medium may be located in an ASIC. The software module may be stored in the memory of the mobile terminal or in a memory card that can be inserted into the mobile terminal. For example, if the device (such as a mobile terminal) uses a large-capacity MEGA-SIM card or a large-capacity flash memory device, the software module may be stored in the MEGA-SIM card or the large-capacity flash memory device.

[0377] One or more of the functional blocks and / or one or more combinations of functional blocks described in the accompanying drawings may be implemented as a 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 discrete gate or transistor logic device, a discrete hardware component, or any appropriate combination thereof for performing the functions described in this application. One or more of the functional blocks and / or one or more combinations of functional blocks described in the accompanying drawings may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in communication with a DSP, or any other such configuration.

[0378] The present application has been described above in conjunction with specific embodiments. However, those skilled in the art should understand that these descriptions are merely illustrative and are not intended to limit the scope of protection of the present application. Those skilled in the art may make various modifications and variations to the present application based on the spirit and principles of the present application, and such modifications and variations are also within the scope of the present application.

[0379] Regarding the above implementation methods disclosed in this embodiment, the following additional notes are also disclosed:

[0380] 1. A data transmission method, wherein the method comprises:

[0381] The network device sends first information to the terminal device, where the first information instructs the terminal device to generate and feed back channel state information;

[0382] The network device receives N layers of channel state information sent by the terminal device, where N is greater than or equal to 2, and channel state information of at least two layers of the N layers of channel state information is indicated by unequal numbers of bits;

[0383] The network device sends data to the terminal device according to the channel state information of the N layer.

[0384] 2. The method according to Supplementary Note 1, wherein the network device sends data to the terminal device according to the channel state information of the N layer, comprising:

[0385] The network device generates a precoding matrix according to the channel state information of the N layers;

[0386] The network device applies the precoding matrix to data sent by the network device to the terminal device;

[0387] The network device sends the data after the precoding matrix is ​​applied to the terminal device.

[0388] 3. The method according to Supplement 1, wherein:

[0389] The channel state information of the N layers is generated based on a codebook, or based on an AL / ML model.

[0390] 4. The method according to Supplement 1, wherein:

[0391] The first information includes a first rule for indicating channel state information of the at least two layers using unequal numbers of bits.

[0392] 5. The method according to Supplementary Note 1, wherein the method further comprises:

[0393] A first rule is predefined or preconfigured for using unequal numbers of bits to indicate the channel state information of the at least two layers.

[0394] 6. The method according to Note 4 or 5, wherein the first rule comprises one or a combination of the following:

[0395] Available AI / ML models;

[0396] Criteria and / or methods for selecting AI / ML models;

[0397] Criteria and / or methods for truncating CSI;

[0398] Approximate method for the number of feedback bits;

[0399] The maximum number of downlink transmission layers;

[0400] Maximum total number of feedback bits;

[0401] The feedback order of CSI.

[0402] 7. The method according to Note 6, wherein the number of available AI / ML models is one or more.

[0403] 8. The method according to Supplementary Note 6, wherein the criterion and / or method for truncating the CSI includes:

[0404] If the length of the CSI bit sequence is greater than a prescribed bit sequence length, truncating the CSI bit sequence;

[0405] If the length of the CSI bit sequence is the same as the length of the prescribed bit sequence, the CSI bit sequence is retained.

[0406] 9. The method according to Supplementary Note 8, wherein the truncating operation on the CSI bit sequence comprises: truncating the CSI bit sequence according to a second rule that is predefined or preconfigured or configured by the network device, wherein the second rule comprises:

[0407] Removing the component at the first position in the CSI bit sequence so that the length of the obtained CSI bit sequence is the same as the length of the specified bit sequence, wherein the first position is one or more bits at any position in the CSI bit sequence, or,

[0408] Perform an IDFT operation on the CSI bit sequence, remove the component at the second position in the CSI bit sequence, and perform a DFT operation on the obtained bit sequence so that the length of the obtained CSI bit sequence is the same as the length of a specified bit sequence, wherein the second position is one or more bits at any position in the CSI bit sequence.

[0409] 10. The method according to Supplementary Note 6, wherein the CSI feedback order includes:

[0410] The order of the CSI values ​​from largest to smallest; or

[0411] The values ​​corresponding to CSI are in ascending order.

[0412] 11. The method according to Note 10, wherein the value corresponding to the CSI includes: an eigenvalue or a singular value of a spatial channel matrix.

[0413] 12. The method according to any one of Notes 6 to 11, wherein the AI / ML model is a bilateral AI / ML model or a unilateral AI / ML model.

[0414] The bilateral AL / ML model refers to an AI / ML model that is set up on both the network device side and the terminal device side;

[0415] The unilateral AL / ML model refers to an AI / ML model set on the network device side or the terminal device side.

[0416] 13. The method according to Supplementary Note 6, wherein there is more than one available AI / ML model, and the method further comprises:

[0417] The network device uses the corresponding AI / ML model to recover the received channel state information according to the available AI / ML model and the criterion and / or method for selecting the AI / ML model.

[0418] 14. The method according to Supplementary Note 6, wherein the available AI / ML model is one, and the method further comprises:

[0419] The network device performs post-processing on the received channel state information based on the first rule;

[0420] The network device restores the channel state information after post-processing.

[0421] 15. The method according to Supplementary Note 6, further comprising:

[0422] The network device receives indication information, where the indication information indicates a number, index, or identifier of an AI / ML model used for the channel state information;

[0423] The network device post-processes the received channel state information according to the number, index, or identifier of the AI / ML model used by the channel state information;

[0424] The network device restores the channel state information after post-processing.

[0425] 16. The method according to Supplement 14 or 15, wherein the post-processing comprises:

[0426] If the length of the CSI bit sequence processed by the terminal device is greater than the prescribed bit sequence length, determining a corresponding padding method according to the criterion and / or method for truncating the CSI included in the first rule, and performing a padding operation on the received CSI bit sequence;

[0427] If the length of the CSI bit sequence processed by the terminal device is the same as the length of the specified bit sequence, the received CSI bit sequence is retained.

[0428] 17. The method according to Supplementary Note 16, wherein performing a lengthening operation on the CSI bit sequence according to the first rule comprises:

[0429] Fill the corresponding bit at the first position with 0 or 1 or a sequence generated according to a predefined rule; or

[0430] An IDFT operation is performed on the bit sequence to be padded, 0 or 1 is added to the corresponding bit position at the second position, or a sequence generated according to a predefined rule is added, and then a DFT operation is performed on the resulting sequence.

[0431] 18. The method according to Note 1, wherein the network device configures the terminal device with the value ranges of the broadband amplitude, the value ranges of the sub-band amplitude, and the value ranges of the phase combining coefficient corresponding to the at least two layers of channel state information, wherein at least one of the value ranges of the broadband amplitude, the value ranges of the sub-band amplitude, and the value ranges of the phase combining coefficient corresponding to the at least two layers of channel state information is different.

[0432] 19. The method according to Note 18, wherein the channel state information includes a PMI, and the PMI includes broadband information (i1) or includes broadband information (i1) and subband information (i2).

[0433] The broadband information (i1) includes:

[0434] The beam offset of the 2D oversampled DFT beam (i 1,1 );

[0435] A beam (i) selected from a set of beams determined according to the beam offset of the two-dimensional oversampled DFT beam 1,2 );

[0436] The index corresponding to the maximum amplitude beam (i 1,3,l );as well as

[0437] Broadband amplitude (i 1,4,l ), the value ranges of the broadband amplitudes corresponding to the channel state information of the at least two layers are the same or different;

[0438] The sub-band information (i2) includes:

[0439] Phase combining coefficient (i 2,1,l ), the value ranges of the phase combining coefficients corresponding to the channel state information of the at least two layers are the same or different; and

[0440] Subband amplitude (i 2,2,l ), the value ranges of the sub-band amplitudes corresponding to the channel state information of the at least two layers are the same or different.

[0441] 20. The method according to Supplementary Note 1, wherein the method further comprises:

[0442] The network device configures a reportQuatity that requires feedback to the terminal device, where the reportQuatity includes at least one of the following: the number of feedback bits of the at least two layers of channel state information, and the index, number, or identifier of the AI / ML model used by each of the at least two layers of channel state information.

[0443] 21. A method for feeding back channel state information, wherein the method comprises:

[0444] The terminal device receives first information sent by the network device, where the first information instructs the terminal device to generate and feed back channel state information;

[0445] The terminal device generates N layers of channel state information, where N is greater than or equal to 2, wherein the channel state information of at least two layers of the N layers of channel state information is indicated by unequal numbers of bits;

[0446] The terminal device sends the channel state information of the N layer to the network device.

[0447] 22. The method according to Note 21, wherein the channel state information of the N layers is generated based on a codebook or based on an AI / ML model.

[0448] 23. The method according to note 21, wherein the terminal device generates the channel state information of the N layers according to a first rule that is predefined or preconfigured or configured by the network device, and the first rule includes one or a combination of the following:

[0449] Available AI / ML models;

[0450] Criteria and / or methods for selecting AI / ML models;

[0451] Criteria and / or methods for truncating CSI;

[0452] Approximate method for the number of feedback bits;

[0453] The maximum number of downlink transmission layers;

[0454] Maximum total number of feedback bits;

[0455] The feedback order of CSI.

[0456] 24. The method according to Note 23, wherein the number of available AI / ML models is one or more.

[0457] 25. The method according to Supplementary Note 23, wherein the criterion and / or method for truncating the CSI comprises:

[0458] If the length of the CSI bit sequence is greater than a prescribed bit sequence length, truncating the CSI bit sequence;

[0459] If the length of the CSI bit sequence is the same as the length of the prescribed bit sequence, the CSI bit sequence is retained.

[0460] 26. The method according to note 25, wherein the truncating operation on the bit sequence of the CSI comprises: truncating the bit sequence of the CSI according to a second rule that is predefined or preconfigured or configured by the network device, wherein the second rule comprises:

[0461] Removing the component at the first position in the CSI bit sequence so that the length of the obtained CSI bit sequence is the same as the length of the specified bit sequence, wherein the first position is one or more bits at any position in the CSI bit sequence, or,

[0462] Perform an IDFT operation on the CSI bit sequence, remove the component at the second position in the CSI bit sequence, and perform a DFT operation on the obtained bit sequence so that the length of the obtained CSI bit sequence is the same as the length of a specified bit sequence, wherein the second position is one or more bits at any position in the CSI bit sequence.

[0463] 27. The method according to Supplementary Note 23, wherein the feedback order of the CSI comprises:

[0464] The order of the CSI values ​​from largest to smallest; or

[0465] The values ​​corresponding to CSI are in ascending order.

[0466] 28. The method according to Note 27, wherein the value corresponding to the CSI includes: an eigenvalue or a singular value of a spatial channel matrix.

[0467] 29. The method according to any one of Notes 23 to 28, wherein the AI / ML model is a bilateral AI / ML model or a unilateral AI / ML model.

[0468] The bilateral AL / ML model refers to an AI / ML model that is set up on both the network device side and the terminal device side;

[0469] The unilateral AL / ML model refers to an AI / ML model set on the network device side or the terminal device side.

[0470] 30. The method according to Supplementary Note 23, further comprising:

[0471] The terminal device sends indication information to the network device, where the indication information indicates the number, index, or identifier of the AI / ML model used by the channel state information, so that the network device post-processes the received channel state information according to the number, index, or identifier of the AI / ML model used by the channel state information, and restores the post-processed channel state information.

[0472] 31. The method according to Supplementary Note 21, further comprising:

[0473] The terminal device receives first configuration information sent by the network device, and the first configuration information is used to configure the value range of the broadband amplitude, the value range of the sub-band amplitude, and the value range of the phase combining coefficient corresponding to the at least two layers of channel state information for the terminal device, wherein at least one of the value range of the broadband amplitude, the value range of the sub-band amplitude, and the value range of the phase combining coefficient corresponding to the at least two layers of channel state information is different.

[0474] 32. The method according to Note 31, wherein the channel state information includes a PMI, and the PMI includes broadband information (i1) or includes broadband information (i1) and subband information (i2).

[0475] The broadband information (i1) includes:

[0476] The beam offset of the 2D oversampled DFT beam (i 1,1 );

[0477] A beam (i) selected from a set of beams determined according to the beam offset of the two-dimensional oversampled DFT beam 1,2 );

[0478] The index corresponding to the maximum amplitude beam (i 1,3,l );as well as

[0479] Broadband amplitude (i 1,4,l ), the value ranges of the broadband amplitudes corresponding to the channel state information of the at least two layers are the same or different;

[0480] The sub-band information (i2) includes:

[0481] Phase combining coefficient (i 2,1,l), the value ranges of the phase combining coefficients corresponding to the channel state information of the at least two layers are the same or different; and

[0482] Subband amplitude (i 2,2,l ), the value ranges of the sub-band amplitudes corresponding to the channel state information of the at least two layers are the same or different.

[0483] 33. The method according to Supplementary Note 21, further comprising:

[0484] The terminal device receives second configuration information sent by the network device, where the second configuration information includes: reportQuatity that requires feedback, and the reportQuatity that requires feedback includes at least one of the following: the number of feedback bits of the at least two layers of channel state information, and the index, number, or identifier of the AI / ML model used by each of the at least two layers of channel state information.

[0485] 34. A network device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program to implement the method as described in any one of Notes 1 to 20.

[0486] 35. A terminal device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program to implement the method as described in any one of Notes 21 to 33.

[0487] 36. A communication system, comprising a terminal device and a network device, wherein:

[0488] The terminal device is configured to execute the method described in any one of Notes 21 to 33, and the network device is configured to execute the method described in any one of Notes 1 to 20.

Claims

1. A data sending device, configured in a network device, wherein: The device comprises: A first sending unit, configured to send first information to a terminal device, where the first information instructs the terminal device to generate and feed back channel state information; a receiving unit configured to receive N layers of channel state information sent by the terminal device, where N is greater than or equal to 2, and channel state information of at least two layers of the N layers of channel state information is indicated by unequal numbers of bits; A second sending unit sends data to the terminal device according to the channel state information of the N layer.

2. The device according to claim 1, wherein The second sending unit generates a precoding matrix according to the channel state information of the N layers; applies the precoding matrix to the data sent by the network device to the terminal device; and sends the data after the precoding matrix is ​​applied to the terminal device.

3. The device according to claim 1, wherein The channel state information of the N layers is generated based on a codebook, or based on an artificial intelligence / machine learning model.

4. The device according to claim 1, wherein The first information includes a first rule for indicating channel state information of the at least two layers using unequal numbers of bits.

5. The device according to claim 4, wherein The first rule is predefined or preconfigured.

6. The device according to claim 4, wherein The first rule includes one or a combination of the following: Available AI / machine learning models; Criteria and / or methods for selecting AI / ML models; Guidelines and / or methods for intercepting channel status information; An approximate method for the number of feedback bits; The maximum number of downlink transmission layers; Maximum total number of feedback bits; Feedback order of channel state information.

7. The device according to claim 6, wherein There is more than one available artificial intelligence / machine learning model, and the second sending unit uses the corresponding artificial intelligence / machine learning model to recover the received channel state information based on the available artificial intelligence / machine learning model and the criteria and / or method for selecting the artificial intelligence / machine learning model.

8. The device according to claim 6, wherein There is one available artificial intelligence / machine learning model, and the second sending unit post-processes the received channel state information based on the first rule; and restores the post-processed channel state information.

9. The device according to claim 6, wherein The receiving unit further receives indication information, where the indication information indicates a number, index, or identifier of an artificial intelligence / machine learning model used by the channel state information; The second sending unit post-processes the received channel state information according to the number, index or identifier of the artificial intelligence / machine learning model used by the channel state information; and restores the post-processed channel state information.

10. The device according to claim 1, wherein The device further comprises: A first configuration unit is configured to configure the terminal device with the value ranges of the broadband amplitude, the value ranges of the sub-band amplitude, and the value ranges of the phase combining coefficient corresponding to the channel state information of the at least two layers, wherein at least one of the value ranges of the broadband amplitude, the value ranges of the sub-band amplitude, and the value ranges of the phase combining coefficient corresponding to the channel state information of the at least two layers is different.

11. The method according to claim 6, wherein: The artificial intelligence / machine learning model is a bilateral artificial intelligence / machine learning model or a unilateral artificial intelligence / machine learning model, The bilateral artificial intelligence / machine learning model refers to an artificial intelligence / machine learning model set up on both the network device side and the terminal device side; The unilateral artificial intelligence / machine learning model refers to an artificial intelligence / machine learning model set on the network device side or the terminal device side.

12. The device according to claim 1, wherein The device further comprises: A second configuration unit configures the terminal device with a report quantity that needs to be fed back, and the report quantity includes at least one of the following: the number of feedback bits of the channel state information of the at least two layers, and the index, number or identifier of the artificial intelligence / machine learning model used by each of the channel state information of the at least two layers.

13. A channel state information feedback device, configured in a terminal device, wherein: The device comprises: a receiving unit configured to receive first information sent by a network device, where the first information instructs the terminal device to generate and feed back channel state information; a generating unit configured to generate N layers of channel state information, where N≥2, wherein channel state information of at least two layers of the N layers of channel state information are indicated by unequal numbers of bits; A sending unit is configured to send the channel state information of the N layer to the network device.

14. The device according to claim 13, wherein The channel state information of the N layers is generated based on a codebook, or based on an artificial intelligence / machine learning model.

15. The device according to claim 13, wherein The generating unit generates the channel state information of the N layers according to a first rule that is predefined or preconfigured or configured by the network device, where the first rule includes one or a combination of the following: Available AI / machine learning models; Criteria and / or methods for selecting AI / ML models; Guidelines and / or methods for intercepting channel status information; An approximate method for the number of feedback bits; The maximum number of downlink transmission layers; Maximum total number of feedback bits; Feedback order of channel state information.

16. The device according to claim 15, wherein The sending unit sends indication information to the network device, where the indication information indicates the number, index, or identifier of the artificial intelligence / machine learning model used by the channel state information, so that the network device post-processes the received channel state information according to the number, index, or identifier of the artificial intelligence / machine learning model used by the channel state information, and restores the post-processed channel state information.

17. The device according to claim 13, wherein The receiving unit receives first configuration information sent by the network device, and the first configuration information is used to configure the value range of the broadband amplitude, the value range of the sub-band amplitude, and the value range of the phase combining coefficient corresponding to the channel state information of the at least two layers for the terminal device, wherein at least one of the value range of the broadband amplitude, the value range of the sub-band amplitude, and the value range of the phase combining coefficient corresponding to the channel state information of the at least two layers is different.

18. The device according to claim 15, wherein The artificial intelligence / machine learning model is a bilateral artificial intelligence / machine learning model or a unilateral artificial intelligence / machine learning model, The bilateral artificial intelligence / machine learning model refers to an artificial intelligence / machine learning model set up on both the network device side and the terminal device side; The unilateral artificial intelligence / machine learning model refers to an artificial intelligence / machine learning model set on the network device side or the terminal device side.

19. The device according to claim 13, wherein The receiving unit receives second configuration information sent by the network device, where the second configuration information includes: the amount of reports that need to be fed back, and the amount of reports that need to be fed back includes at least one of the following: the number of feedback bits of the at least two layers of channel state information, and the index, number, or identifier of the artificial intelligence / machine learning model used by each of the at least two layers of channel state information.

20. A communication system comprising a terminal device and a network device, wherein: The network device is configured to: Sending first information to the terminal device, where the first information instructs the terminal device to generate and feed back channel state information; Receiving N layers of channel state information sent by a terminal device, where N is greater than or equal to 2, and channel state information of at least two layers of the N layers of channel state information is indicated by unequal numbers of bits; Sending data to the terminal device according to the channel state information of the N layer; The terminal device is configured as follows: receiving the first information sent by the network device; Generating channel state information of the N layers; The channel state information of the N layer is sent to the network device.

Images