Channel characteristic information transmission method and apparatus, terminal, and network-side device

Abstract

The application discloses a channel feature information transmission method and device, a terminal and a network side equipment, and belongs to the technical field of communication. The channel feature information transmission method comprises the following steps: inputting channel information into a first artificial intelligence (AI) network model and a second AI network model by a terminal, and acquiring first channel feature information output by the first AI network model and second channel feature information output by the second AI network model; and reporting at least one of the first channel feature information and the second channel feature information to a network side equipment by the terminal.

Description

Technical Field

[0001] This application belongs to the field of communication technology, and specifically relates to a method, apparatus, terminal and network-side equipment for transmitting channel characteristic information. Background Technology

[0002] With the development of science and technology, people have begun to study the application of Artificial Intelligence (AI) networks in communication systems. For example, communication data can be transmitted between network-side devices and terminals through AI network models. Currently, the AI ​​network model used for encoding on the terminal side and the AI ​​network model used for decoding on the network side are jointly trained. During training, the encoded bits do not have errors. However, during air interface transmission, if the channel information is large, channel characteristic information may be lost. Summary of the Invention

[0003] This application provides a method, apparatus, terminal, and network-side device for transmitting channel feature information, which can solve the problem of large channel information that may be lost in related technologies.

[0004] Firstly, a method for transmitting channel characteristic information is provided, including:

[0005] The terminal inputs channel information into the first AI network model and the second AI network model, and obtains the first channel feature information output by the first AI network model and the second channel feature information output by the second AI network model.

[0006] The terminal reports at least one of the first channel feature information and the second channel feature information to the network-side device.

[0007] Secondly, a method for transmitting channel characteristic information is provided, including:

[0008] The network-side device receives at least one of the first channel characteristic information and the second channel characteristic information reported by the terminal;

[0009] Wherein, the first channel feature information is output by the terminal through the first AI network model, and the second channel feature information is output by the terminal through the second AI network model.

[0010] Thirdly, a channel feature information transmission device is provided, comprising:

[0011] The acquisition module is used to input channel information into the first artificial intelligence (AI) network model and the second AI network model, and to acquire the first channel feature information output by the first AI network model and the second channel feature information output by the second AI network model.

[0012] The reporting module is used to report at least one of the first channel feature information and the second channel feature information to the network-side device.

[0013] Fourthly, a channel feature information transmission device is provided, comprising:

[0014] A receiving module is configured to receive at least one of the first channel characteristic information and the second channel characteristic information reported by the terminal.

[0015] Wherein, the first channel feature information is output by the terminal through the first AI network model, and the second channel feature information is output by the terminal through the second AI network model.

[0016] Fifthly, a terminal is provided, comprising a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the channel feature information transmission method as described in the first aspect.

[0017] In a sixth aspect, a terminal is provided, including a processor and a communication interface, wherein the processor is used to input channel information into a first artificial intelligence (AI) network model and a second AI network model, and to obtain first channel feature information output by the first AI network model and second channel feature information output by the second AI network model, and the communication interface is used to report at least one of the first channel feature information and the second channel feature information to a network-side device.

[0018] In a seventh aspect, a network-side device is provided, the network-side device including a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the channel feature information transmission method as described in the second aspect.

[0019] Eighthly, a network-side device is provided, including a processor and a communication interface, wherein the communication interface is used to receive at least one of a first channel feature information and a second channel feature information reported by a terminal; wherein the first channel feature information is output by the terminal through a first AI network model, and the second channel feature information is output by the terminal through a second AI network model.

[0020] A ninth aspect provides a communication system comprising: a terminal and a network-side device, wherein the terminal is configured to perform the steps of the channel feature information transmission method as described in the first aspect, and the network-side device is configured to perform the steps of the channel feature information transmission method as described in the second aspect.

[0021] In a tenth aspect, a readable storage medium is provided, on which a program or instructions are stored, which, when executed by a processor, implement the steps of the channel feature information transmission method as described in the first aspect, or implement the steps of the channel feature information transmission method as described in the second aspect.

[0022] Eleventhly, a chip is provided, the chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the channel feature information transmission method as described in the first aspect, or to implement the channel feature information transmission method as described in the second aspect.

[0023] In a twelfth aspect, a computer program / program product is provided, which is stored in a storage medium and executed by at least one processor to implement the channel feature information transmission method as described in the first aspect, or to implement the channel feature information transmission method as described in the second aspect.

[0024] In this embodiment, the terminal can input channel information into a first AI network model and a second AI network model respectively, obtain first channel feature information output by the first AI network model and second channel feature information output by the second AI network model, and report this information to the network-side device. The network-side device can then process the channel information using corresponding third and fourth AI network models to recover the channel information. Thus, for long channel information, the terminal and network-side device can process the channel information using two sets of AI network models, with each set processing a portion of the channel information, avoiding transmission errors and loss of channel information. Attached Figure Description

[0025] Figure 1 This is a block diagram of a wireless communication system applicable to embodiments of this application;

[0026] Figure 2 This is a flowchart of a channel feature information transmission method provided in an embodiment of this application;

[0027] Figure 2a This is one of the schematic diagrams illustrating the preprocessing of channel information in a channel feature information transmission method provided in this application embodiment;

[0028] Figure 2b This is a second schematic diagram illustrating the preprocessing of channel information in a channel feature information transmission method provided in this application embodiment;

[0029] Figure 2c This is the third schematic diagram of the channel information preprocessing method provided in the embodiment of this application for transmitting channel feature information;

[0030] Figure 2d This is one of the schematic diagrams of the splicing operation in a channel feature information transmission method provided in this application embodiment;

[0031] Figure 2e This is a second schematic diagram of the splicing operation in a channel feature information transmission method provided in this application embodiment;

[0032] Figure 2f This is the third schematic diagram of the splicing operation in a channel feature information transmission method provided in this application embodiment;

[0033] Figure 2g This is the fourth schematic diagram of the splicing operation in a channel feature information transmission method provided in this application embodiment;

[0034] Figure 3 This is a flowchart of another channel feature information transmission method provided in the embodiments of this application;

[0035] Figure 4 This is a structural diagram of a channel feature information transmission device provided in an embodiment of this application;

[0036] Figure 5 This is a structural diagram of another channel feature information transmission device provided in the embodiments of this application;

[0037] Figure 6 This is a structural diagram of a communication device provided in an embodiment of this application;

[0038] Figure 7 This is a structural diagram of a terminal provided in an embodiment of this application;

[0039] Figure 8 This is a structural diagram of a network-side device provided in an embodiment of this application. Detailed Implementation

[0040] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0041] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0042] It is worth noting that the technologies described in this application are not limited to Long Term Evolution (LTE) / LTE-Advanced (LTE-A) systems, but can also be used in other wireless communication systems, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single-carrier Frequency Division Multiple Access (SC-FDMA), and other systems. The terms "system" and "network" in this application are often used interchangeably, and the described technologies can be used with the systems and radio technologies mentioned above, as well as with other systems and radio technologies. The following description describes New Radio (NR) systems for illustrative purposes, and NR terminology is used in most of the following description; however, these technologies can also be applied to applications beyond NR systems, such as 6th generation (6G) radio systems. th Generation 6G communication system.

[0043] Figure 1This diagram illustrates a block diagram of a wireless communication system applicable to embodiments of this application. The wireless communication system includes a terminal 11 and a network-side device 12. Terminal 11 can be a mobile phone, tablet computer, laptop computer, personal digital assistant (PDA), handheld computer, netbook, ultra-mobile personal computer (UMPC), mobile internet device (MID), augmented reality (AR) / virtual reality (VR) device, robot, wearable device, vehicle-mounted device (VUE), pedestrian terminal (PUE), smart home (home devices with wireless communication capabilities, such as refrigerators, televisions, washing machines, or furniture), game console, personal computer (PC), ATM, or self-service machine, etc. Wearable devices include: smartwatches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart chains, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. It should be noted that the specific type of terminal 11 is not limited in this embodiment. Network-side equipment 12 may include access network equipment or core network equipment. Access network equipment may also be referred to as radio access network equipment, radio access network (RAN), radio access network function, or radio access network unit. Access network equipment may include base stations, WLAN access points, or WiFi nodes, etc. Base stations may be referred to as Node B, evolved Node B (eNB), access point, base transceiver station (BTS), radio base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), home B node, home evolved B node, Transmitting Receiving Point (TRP), or any other suitable term in the field, as long as the same technical effect is achieved. The base station is not limited to specific technical terms. It should be noted that this application embodiment only uses a base station in an NR system as an example for description and does not limit the specific type of base station.

[0044] To better understand the technical solution of this application, the relevant concepts that may be involved in the embodiments of this application are explained below.

[0045] As information theory dictates, accurate channel state information (CSI) is crucial for channel capacity. This is especially true for multi-antenna systems, where the transmitter can optimize signal transmission based on CSI to better match the channel state. For example, channel quality indicator (CQI) can be used to select a suitable modulation and coding scheme (MCS) for link adaptation; precoding matrix indicator (PMI) can be used to implement eigenbeamforming to maximize received signal strength or to suppress interference (such as inter-cell interference or multi-user interference). Therefore, since the introduction of multi-input multi-output (MIMO) technology, CSI acquisition has been a hot research topic.

[0046] Typically, network-side equipment (such as base stations) transmits CSI (channel state information reference signal) on certain time-frequency resources in a slot. The terminal performs channel estimation based on the CSI-RS, calculates the channel information in that slot, and feeds back the PMI (Programmable Memory Information) to the base station through a codebook. The base station combines the channel information based on the codebook information fed back by the terminal. Before the next CSI report, the base station uses this information for data precoding and multi-user scheduling.

[0047] To further reduce CSI feedback overhead, the terminal can change the PMI reporting for each subband to reporting PMI according to delay. Since the channel in the delay domain is more concentrated, the PMI of all subbands can be approximately represented by PMI with less delay. That is, the delay domain information is compressed before reporting.

[0048] Similarly, to reduce overhead, the base station can pre-encode the CSI-RS and send the encoded CSI-RS to the terminal. The terminal sees the channel corresponding to the encoded CSI-RS. The terminal only needs to select a few ports with higher strength from the ports indicated by the network side and report the coefficients corresponding to these ports.

[0049] Furthermore, to better compress channel information, neural networks or machine learning methods can be used. Specifically, the terminal uses an AI network model to compress and encode the channel information, and the base station uses an AI network model to decode the compressed content to recover the channel information. At this point, the AI ​​network model used for decoding at the base station and the AI ​​network model used for encoding at the terminal need to be jointly trained to achieve a reasonable degree of matching. A joint neural network model is formed by the AI ​​network model used for encoding at the terminal and the AI ​​network model used for decoding at the base station, and this model is jointly trained by the network side. After training, the base station sends the AI ​​network model used for encoding to the terminal.

[0050] The terminal estimates CSI-RS, calculates channel information, and uses the calculated channel information or the original estimated channel information to obtain the encoding result through the AI ​​network model. The encoding result is then sent to the base station, which receives the encoded result and inputs it into the AI ​​network model for decoding to recover the channel information.

[0051] Different channel environments result in varying degrees of compressibility and encoding of channel information, leading to different encoded information lengths. For a given compression performance target, simple channel information requires only a short encoding length, while complex channel information necessitates a longer encoding length. Generally, different encoded information lengths correspond to different AI network models, thus requiring multiple AI network models to handle different compressed bit lengths. This necessitates the training and configuration of multiple AI network models on both the terminal and network sides.

[0052] The channel feature information transmission method provided in this application will be described in detail below with reference to the accompanying drawings and through some embodiments and application scenarios.

[0053] Please refer to Figure 2 , Figure 2 This is a flowchart of a channel feature information transmission method provided in an embodiment of this application, wherein the execution subject of the method is a terminal. Figure 2 As shown, the method includes the following steps:

[0054] Step 201: The terminal inputs the channel information into the first AI network model and the second AI network model, and obtains the first channel feature information output by the first AI network model and the second channel feature information output by the second AI network model.

[0055] Optionally, the terminal may detect Channel State Information Reference Signal (CSI-RS) or Tracking Reference Signal (TRS) at a designated location on the network side device, perform channel estimation, and obtain channel information. The terminal then inputs this channel information into a first AI network model and a second AI network model. The first AI network model compresses and encodes the input channel information (such as the channel matrix for each sub-band or the precoding matrix for each sub-band) and outputs first channel feature information. The second AI network model compresses and encodes the channel information and outputs second channel feature information. The first and second channel feature information may be Channel State Information (CSI) or other information related to channel information. It should be noted that the channel information encoding mentioned in this embodiment is different from channel coding.

[0056] Step 202: The terminal reports at least one of the first channel feature information and the second channel feature information to the network-side device.

[0057] For example, the terminal reports first channel feature information and second channel feature information to the network-side device, or it may only report the first channel feature information. Based on the received first channel feature information and / or second channel feature information, the network-side device inputs the first channel feature information and / or second channel feature information into the corresponding AI network model for decoding processing to recover the channel information.

[0058] It should be noted that the network-side device includes a third AI network model corresponding to the first AI network model and a fourth AI network model corresponding to the second AI network model. The network-side device jointly trains the first AI network model and the third AI network model, and jointly trains the second AI network model and the fourth AI network model, so that the first AI network model matches the third AI network model, and the second AI network model matches the fourth AI network model. The output of the first AI network model is used as the input of the third AI network model, and the output of the second AI network model is used as the input of the fourth AI network model. The output of the third AI network model matches the input of the first AI network model, and the output of the fourth AI network model matches the input of the second AI network model. In this way, the channel information transmission between the terminal and the network-side device can be realized through the processing of channel information by the AI ​​network model.

[0059] In this embodiment, the terminal reports the first channel feature information output by the first AI network model to the network-side device. The network-side device then inputs the first channel feature information into the third AI network model, which decodes the first channel feature information to obtain partial channel information. Similarly, the terminal reports the second channel feature information output by the second AI network model to the network-side device, which inputs the second channel feature information into the fourth AI network model. The fourth AI network model decodes the second channel feature information to obtain another portion of channel information. The network-side device combines the partial channel information output by the third AI network model with the other portion of channel information output by the fourth AI network model to recover the complete channel information, which is the same channel information input into the first and second AI network models. In this way, the complete channel information can be recovered using the third and fourth AI network models of the network-side device. For longer channel information, the terminal and network-side device can process the channel information using two sets of AI network models, with each set processing a portion of the channel information, thus avoiding transmission errors and loss of channel information.

[0060] Optionally, the length of the first channel feature information is a fixed value, and the length of the second channel feature information is either a fixed value or a variable value. That is, the first AI network model encodes the channel information into a fixed-length first channel feature information, while the length of the second channel feature information obtained by the second AI network model is not fixed; for example, it can be determined based on the length of the first channel feature information. Correspondingly, the length of the first channel feature information input to the third AI network model is fixed, and the channel information decoded by the third AI network model can also be specific. However, the length of the second channel feature information input to the fourth AI network model is not fixed, therefore the length of the channel information output by the fourth AI network model is also not fixed. The terminal and network-side devices can pre-determine the length of the first channel feature information. Then, for channel information of different lengths, a fixed length of channel information can be recovered through encoding by the first AI network model and decoding by the third AI network model. For example, the first channel feature information can be the more important channel information, thus avoiding the loss of important channel information. In addition, for channel information of different lengths, the terminal and network-side devices can pre-determine the length of the corresponding second channel feature information for channel information of different lengths. Through encoding by the second AI network model and decoding by the fourth AI network model, channel information of different lengths can be recovered. This allows for more flexible encoding and decoding of channel information of different lengths.

[0061] Optionally, the second channel feature information can also be a fixed value, for example, the length of the first channel feature information is the same as the length of the second channel feature information. Then, the network-side device can recover half of the channel information using a third AI network model and the other half using a fourth AI network model, combining them to obtain the complete channel information. The half of the channel information can be channel information from subbands 1, 3, 5, and 7, or channel information from subbands 1, 2, 3, and 4, or channel information from a single polarization, etc.

[0062] It should be noted that in this embodiment, the terminal may directly input the channel information into the first AI network model and the second AI network model without performing any processing. For example, please refer to... Figure 2a In a beam-delay network, the terminal can either perform no processing on the channel information, inputting the channel matrix of each sub-band into a first AI network model for beam coding and into a second AI network model for delay coding, and then concatenating the output channel feature information. Alternatively, the terminal can preprocess the channel information before inputting it into the first and second AI network models.

[0063] Optionally, the terminal inputs channel information into the first AI network model and the second AI network model, including at least one of the following:

[0064] The terminal inputs the channel information into the first AI network model after first preprocessing.

[0065] The terminal inputs the channel information into the second AI network model after undergoing a second preprocessing step.

[0066] For example, the terminal may perform a first preprocessing on the channel information before inputting it into the first AI network model. For instance, in a wideband subband network, the first preprocessing may be calculating the sum of the second moments of the channel matrices of all subbands; or, in a beam-delay network, the first preprocessing may be projecting the channel information onto an orthogonal discrete Fourier transform (DFT) basis; and so on.

[0067] Optionally, inputting the channel information into the second AI network model after second preprocessing includes any one of the following:

[0068] The channel information is input into the first AI network model, and the output of the first AI network model is input into the second AI network model;

[0069] The channel information is input into the first AI network model, and the output of the target network structure in the first AI network model is input into the second AI network model.

[0070] For example, the terminal inputs channel information into the first AI network model, uses the output of the first AI network as the input of the second AI network model, or uses the output of a certain network structure in the first AI network model as the input of the second AI network model, such as... Figure 2b As shown, the output of the first AI network model before encoding and quantization is used as the second AI network model. Figure 2b The input (neutron band coding) is used to perform a second preprocessing of the channel information.

[0071] Alternatively, the second preprocessing can be the same as the first preprocessing, meaning the terminal can input the channel information, after the first preprocessing, into both the first and second AI network models, such as... Figure 2c As shown, for example, the second moment of the channel matrix of all sub-bands is calculated and then input into the second AI network model. Figure 2c Neutron band coding).

[0072] In this embodiment, the channel information input to the first AI network model may undergo first preprocessing only, or it may undergo second preprocessing only, or it may be directly input into the first and second AI network models without any preprocessing. Alternatively, it may involve first preprocessing of the channel information input to the first AI network model and second preprocessing of the channel information input to the second AI network model. This allows the terminal to process the channel information in different ways depending on the channel information conditions, making the terminal's channel information processing more flexible.

[0073] In this embodiment of the application, the first channel feature information and the second channel feature information may be unrelated to CSI. In this case, the first channel feature information and the second channel feature information may be reported to the network-side device independently by the terminal; or, the first channel feature information and the second channel feature information may also be reported through CSI.

[0074] Optionally, when the first channel feature information and the second channel feature information are reported via CSI, the terminal reports at least one of the first channel feature information and the second channel feature information to the network-side device, including any one of the following:

[0075] The terminal reports the first channel feature information to the network-side device through the first CSI, and reports the second channel feature information to the network-side device through the second CSI;

[0076] The terminal reports the first channel feature information and the second channel feature information to the network-side device through the first CSI, wherein the first channel feature information is contained in the first part of the first CSI and the second channel feature information is contained in the second part of the first CSI.

[0077] In one implementation, the terminal reports the first channel feature information and the second channel feature information respectively through two CSIs. For example, the first channel feature information is carried in the first CSI and the second channel feature information is carried in the second CSI.

[0078] In another implementation, the terminal may report the first channel feature information and the second channel feature information through a CSI. For example, the first channel feature information is carried in a first part (CSIPart1) of the first CSI, and the second channel feature information is carried in a second part (CSIPart2) of the first CSI. The first part is a fixed-length portion of the first CSI, and the second part is a variable-length portion of the first CSI.

[0079] Optionally, the first part (CSI Part 1) may also include the length of the second channel feature information. The network-side device may directly obtain the length of the second channel feature information from CSI Part 1 and decode the first and second channel feature information in parallel. In this way, the network-side device can more accurately decode the second channel feature information based on its length.

[0080] In this embodiment, the terminal may report only the first channel characteristic information to the network-side device, or it may report both the first channel characteristic information and the second channel characteristic information. Optionally, the terminal reporting at least one of the first channel characteristic information and the second channel characteristic information to the network-side device includes:

[0081] The terminal concatenates the first channel feature information and the second channel feature information to obtain the target channel feature information;

[0082] The terminal reports the target channel characteristic information to the network-side device.

[0083] In other words, the terminal concatenates the first channel feature information and the second channel feature information, for example, by concatenating the second channel feature information after the first channel feature information, and then reports the concatenated first channel feature information and second channel feature information to the network-side device.

[0084] Optionally, the terminal concatenates the first channel feature information and the second channel feature information to obtain target channel feature information, including any one of the following:

[0085] The terminal concatenates the second channel feature information after the first channel feature information to obtain the target channel feature information;

[0086] The terminal interleaves and splices the first channel feature information and the second channel feature information to obtain the target channel feature information;

[0087] The terminal concatenates the second channel feature information after the first channel feature information, and performs a first operation on the concatenated first channel feature information and second channel feature information with a preset scrambling code sequence to obtain target channel feature information;

[0088] The terminal performs a second operation on the second channel feature information and the first channel feature information to obtain the third channel feature information, and then concatenates the third channel feature information after the first channel feature information to obtain the target channel feature information.

[0089] Please refer to Figure 2d In the diagram, A represents the first channel feature information, and B represents the second channel feature information. The terminal concatenates the second channel feature information with the first channel feature information to obtain the target channel feature information (i.e., the concatenated first and second channel feature information). The terminal reports the target channel feature information to the network-side device. The network-side device can use a corresponding method to decode the target channel feature information into the first and second channel feature information, and then input them into the third and fourth AI network models respectively for decoding processing.

[0090] Please refer to Figure 2e In the diagram, A represents the first channel feature information, and B represents the second channel feature information. The terminal interleaves and splices the first and second channel feature information to obtain the target channel feature information. The interleaving and splicing can involve randomly inserting the second channel feature information into the first channel feature information. Optionally, the first channel feature information may include the same identifier, and the second channel feature information may also include the same identifier. The network-side device can then decode the target channel feature information into the first and second channel feature information by recognizing the identifiers, and then input them into the third and fourth AI network models respectively for decoding processing.

[0091] Please refer to Figure 2fIn the diagram, A represents the first channel feature information, and B represents the second channel feature information. The terminal concatenates the second channel feature information after the first channel feature information. Then, it multiplies the concatenated first and second channel feature information with a preset scrambling sequence to obtain the target channel feature information. The preset scrambling sequence can be configured by the network-side device. After receiving the target channel feature information reported by the terminal, the network-side device processes the target channel feature information based on the preset scrambling sequence to obtain the first and second channel feature information, which are then input into the third and fourth AI network models for decoding, respectively.

[0092] Optionally, the length of the first channel feature information is related to the preset scrambling code sequence. For example, the relationship between the length of the first channel feature information and the preset scrambling code sequence may be a network-side device configuration or a protocol agreement.

[0093] Please refer to Figure 2g In the diagram, A represents the first channel feature information, and B represents the second channel feature information. The terminal can calculate the difference between the second and first channel feature information to obtain the third channel feature information, then concatenate the third channel feature information after the first channel feature information to obtain the target channel feature information, which is then reported. The network-side equipment correspondingly processes the target channel feature information to obtain the first and second channel feature information, which are then input into the third and fourth AI network models for decoding.

[0094] Optionally, the terminal reports the target channel characteristic information to the network-side device, including:

[0095] The terminal quantizes the target channel feature information through a quantization network and reports the quantized target channel feature information to the network-side device.

[0096] The quantization network is an AI quantization network model. The terminal concatenates the first channel feature information and the second channel feature information, then quantizes the concatenated information together through the quantization network before reporting it to the network-side device. Optionally, the first channel feature information can be output after quantization processing by a first AI network model, and the second channel feature information can be output after quantization processing by a second AI network model. That is, the first and second channel feature information have already undergone one quantization before concatenation, and the target channel feature information obtained after concatenation can be quantized again through the quantization network. In this embodiment, quantizing the channel feature information effectively compresses and simplifies its capacity.

[0097] Optionally, obtaining the first channel feature information output by the first AI network model and the second channel feature information output by the second AI network model includes:

[0098] Obtain the first channel feature information output by the first AI network model before quantization and the second channel feature information output by the second AI network model before quantization;

[0099] The terminal reports at least one of the first channel feature information and the second channel feature information to the network-side device, including:

[0100] The terminal concatenates the first channel feature information and the second channel feature information to obtain the target channel feature information;

[0101] The terminal quantizes the target channel feature information through a quantization network and reports the quantized target channel feature information to the network-side device.

[0102] In this embodiment, the terminal concatenates the first channel feature information output before the first AI network model is quantized and the second channel feature information output before the second AI network model is quantized. Then, the concatenated first and second channel feature information are quantized, and the quantized first and second channel feature information are reported to the network-side device to compress and simplify the capacity of the channel feature information.

[0103] In this embodiment of the application, the terminal reports at least one of the first channel feature information and the second channel feature information to the network-side device, including:

[0104] The terminal reports the first channel feature information to the network-side device and discards the second channel feature information.

[0105] It should be noted that the terminal may discard the second channel feature information during the process of reporting the first and second channel feature information to the network-side device; or, the terminal may discard the second channel feature information before reporting, that is, only report the first channel feature information. In this embodiment, for example, when channel resources are insufficient, the terminal discards the second channel feature information and reports only the first channel feature information to ensure that the network-side device can at least receive the first channel feature information and decode it through the third AI network model to recover part of the channel information and avoid the complete loss of channel information.

[0106] Optionally, the terminal includes a first AI network model and at least one second AI network model, and the terminal inputs channel information into the first AI network model and the second AI network model, including:

[0107] The terminal inputs channel information into the first AI network model and the target second AI network model, wherein the at least one second AI network model includes the target second AI network model:

[0108] The target second AI network model is determined by at least one of the following:

[0109] Channel environment;

[0110] Instructions for the network-side devices.

[0111] It should be noted that the first AI network model and at least one second AI network model are trained by the network-side device and then sent to the terminal. The network-side device may send instruction information to the terminal to indicate which second AI network model the terminal should use, or the terminal may adaptively select a suitable second AI network model as the target second AI network model based on the channel environment. This makes the terminal's selection of the second AI network model more flexible.

[0112] Furthermore, the terminal inputs the channel information into the first AI network model to obtain the first channel feature information, and inputs the channel information into the target second AI network model to obtain the second channel feature information. The terminal may report the lengths of the first and second channel feature information in CSI Part 1, and report the second channel feature information in CSI Part 2.

[0113] Optionally, the first AI network model corresponds to the third AI network model of the network-side device, and the second AI network model corresponds to the fourth AI network model of the network-side device.

[0114] The input to the third AI network model is the first channel feature information, and the input to the fourth AI network model includes any one of the following:

[0115] The second channel feature information;

[0116] The first channel feature information and the second channel feature information;

[0117] The output of the third AI network model and the second channel feature information.

[0118] In this embodiment, the first and third AI network models are jointly trained using a network-side device, as are the second and fourth AI network models. The input of the third AI network model is the output of the first AI network model. The input of the fourth AI network model can be the output of the second AI network model, or it can include the outputs of both the first and second AI network models, or it can include both the outputs of the third and second AI network models. This allows the third AI network model to decode independently based on the first channel feature information, while the fourth AI network model can decode solely based on the second feature information or rely on the first channel feature information. For example, if the first channel feature information is broadband information and the second channel feature information is subband information, the fourth AI network model cannot decode solely based on the subband information and must rely on the broadband information. This allows the network-side device to employ different decoding methods based on the channel feature information, making the decoding of channel feature information more flexible.

[0119] It should be noted that the network-side device decodes the first channel feature information separately through the third AI network model. For positions that are not present in the first channel feature information, zeros are padded or other agreed-upon values ​​are added to ensure that the third AI network model can output the recovered channel information.

[0120] Optionally, the second channel feature information includes N information blocks, which are arranged in a preset order, where N is an integer greater than 1;

[0121] When the fourth AI network model decodes the second channel feature information, the (i+1)th information block among the N information blocks is decoded based on the ith information block, where i is an integer less than N.

[0122] In this embodiment of the application, the second channel feature information can be divided into N information blocks. When the fourth AI network model decodes these N information blocks, each information block depends on the previous information block for decoding, or these N information blocks can be decoded individually, so as to flexibly realize the decoding processing of the second channel feature information.

[0123] Optionally, when uploading the first channel feature information and the second channel feature information, the terminal may discard the second channel feature information. In the case of discarding the second channel feature information, the terminal discards the N information blocks in reverse order of the preset order, that is, discards the N information blocks from back to front according to the preset order. The number of discarded blocks can be determined by the terminal based on channel resources. Furthermore, only a portion of the second channel feature information may be discarded; the remaining portion can still be reported to the network-side device. When the network-side device decodes the received second channel feature information based on the fourth AI network model, it pads the discarded portion with 0s or other agreed-upon values ​​to ensure that the channel information can be decoded and recovered.

[0124] In this embodiment, the network-side device can update the first AI network model and the second AI network model. For example, it can update both AI network models separately, or only update the second AI network model, and then send the updated AI network model to the terminal. Correspondingly, the third AI network model needs to be updated synchronously with the first AI network model, and the fourth AI network model needs to be updated synchronously with the second AI network model to ensure that the terminal and the network-side device can encode and decode the channel information through the corresponding AI network models.

[0125] It should be noted that when the network-side device updates the second AI network model, it can adjust the length of the output of the second AI network model, that is, the length of the second channel feature information. In other words, the length of the second channel feature information is a variable value, which can be selected by the user. This allows for flexible setting of different lengths of the second channel feature information for different channel information.

[0126] Please refer to Figure 3 , Figure 3 This is a flowchart of another channel feature information transmission method provided in this application embodiment, where the execution subject of this method is a network-side device. Figure 3 As shown, the method includes the following steps:

[0127] Step 301: The network-side device receives at least one of the first channel characteristic information and the second channel characteristic information reported by the terminal.

[0128] Wherein, the first channel feature information is output by the terminal through the first AI network model, and the second channel feature information is output by the terminal through the second AI network model.

[0129] In this embodiment, the network-side device includes a third AI network model corresponding to the first AI network model and a fourth AI network model corresponding to the second AI network model. The terminal reports the first channel feature information output by the first AI network model to the network-side device. The network-side device then inputs the first channel feature information into the third AI network model, which decodes the first channel feature information to obtain partial channel information. Similarly, the terminal reports the second channel feature information output by the second AI network model to the network-side device, which then inputs the second channel feature information into the fourth AI network model. The fourth AI network model decodes the second channel feature information to obtain another portion of channel information. The network-side device combines the partial channel information output by the third AI network model with the other portion of channel information output by the fourth AI network model to recover the complete channel information, which is also the channel information input to the first and second AI network models. In this way, the complete channel information can be recovered through the third and fourth AI network models of the network-side device. For long channel information, the terminal and the network-side device can process the channel information through two sets of AI network models respectively. Each set of AI network models processes a part of the channel information, avoiding transmission errors and loss of channel information.

[0130] In this embodiment, the length of the first channel feature information is a fixed value, and the length of the second channel feature information is either a fixed value or a variable value. That is, the first AI network model encodes the channel information into a fixed-length first channel feature information, while the length of the second channel feature information obtained by the second AI network model is not fixed; for example, it can be determined based on the length of the first channel feature information. The terminal and network-side devices can pre-agree on the length of the first channel feature information. Therefore, for channel information of different lengths, the fixed-length channel information can be recovered through encoding by the first AI network model and decoding by the third AI network model. For example, the first channel feature information can be a relatively important channel information, thus avoiding the loss of important channel information. Furthermore, for channel information of different lengths, the terminal and network-side devices can agree on the corresponding length of the second channel feature information for each length. Through encoding by the second AI network model and decoding by the fourth AI network model, channel information of different lengths can be recovered, thus enabling more flexible encoding and decoding of channel information of different lengths.

[0131] Optionally, when the first channel feature information and the second channel feature information are reported via CSI, the network-side device receives at least one of the first channel feature information and the second channel feature information reported by the terminal, including any one of the following:

[0132] The network-side device receives the first channel characteristic information reported by the terminal through the first CSI, and the second channel characteristic information reported through the second CSI;

[0133] The network-side device receives first channel characteristic information and second channel characteristic information reported by the terminal through the first CSI, wherein the first channel characteristic information is contained in the first part of the first CSI and the second channel characteristic information is contained in the second part of the first CSI.

[0134] Optionally, the first part may also include the length of the second channel feature information.

[0135] It should be noted that the specific procedures for the above steps can be referred to Figure 2 The specific descriptions of the method embodiments described herein will not be repeated in this embodiment.

[0136] Optionally, the network-side device receives at least one of the first channel characteristic information and the second channel characteristic information reported by the terminal, including:

[0137] The network-side device receives target channel feature information reported by the terminal. The target channel feature information is channel feature information obtained by the terminal by concatenating the first channel feature information and the second channel feature information.

[0138] The specific implementation process for the terminal to concatenate the first channel feature information and the second channel feature information can be referred to Figure 2 The specific descriptions of the method embodiments described herein will not be repeated in this embodiment.

[0139] Optionally, the network-side device includes a third AI decoding network model corresponding to the first AI network model and a fourth AI network model corresponding to the second AI network model;

[0140] The input to the third AI network model is the first channel feature information, and the input to the fourth AI network model includes any one of the following:

[0141] The second channel feature information;

[0142] The first channel feature information and the second channel feature information;

[0143] The output of the first AI decoding network model and the second channel feature information.

[0144] Optionally, the second channel feature information includes N information blocks, which are arranged in a preset order, where N is an integer greater than 1;

[0145] The method further includes:

[0146] The network-side device decodes the second channel feature information through the fourth AI network model, wherein the (i+1)th information block among the N information blocks is decoded based on the ith information block, where i is an integer less than N.

[0147] Optionally, the method further includes:

[0148] The network-side device updates the first AI network model and the second AI network model; or...

[0149] The network-side device updates the second AI network model.

[0150] Optionally, updating the second AI network model includes adjusting the length of the second channel feature information.

[0151] In this embodiment, the network-side device can update the first AI network model and the second AI network model. For example, it can update both AI network models separately, or only update the second AI network model, and then send the updated AI network model to the terminal. Correspondingly, the third AI network model needs to be updated synchronously with the first AI network model, and the fourth AI network model needs to be updated synchronously with the second AI network model to ensure that the terminal and the network-side device can encode and decode the channel information through the corresponding AI network models.

[0152] Optionally, when updating the second AI network model, the network-side device can adjust the length of the output of the second AI network model, that is, the length of the second channel feature information. In other words, the length of the second channel feature information is a variable value, which can be selected by the user, thereby flexibly setting different lengths of the second channel feature information for different channel information.

[0153] The channel feature information method provided in this application is applied to network-side devices, and the relevant concepts and specific implementation processes involved can be referred to the above. Figure 2 The specific description of the channel feature information transmission method embodiment applied to the terminal will not be repeated in this embodiment to avoid repetition.

[0154] The channel feature information transmission method provided in this application can be executed by a channel feature information transmission device. This application uses the example of a channel feature information transmission device executing the channel feature information transmission method to illustrate the channel feature information transmission device provided in this application.

[0155] Please refer to Figure 4 , Figure 4 This application provides a channel feature information transmission device, such as... Figure 4 As shown, the channel feature information transmission device 400 includes:

[0156] The acquisition module 401 is used to input channel information into the first artificial intelligence (AI) network model and the second AI network model, and to acquire the first channel feature information output by the first AI network model and the second channel feature information output by the second AI network model.

[0157] The reporting module 402 is used to report at least one of the first channel feature information and the second channel feature information to the network-side device.

[0158] Optionally, the acquisition module 401 is further configured to perform at least one of the following:

[0159] The channel information is input into the first AI network model after undergoing the first preprocessing.

[0160] The channel information is then preprocessed and input into the second AI network model.

[0161] Optionally, the acquisition module 401 is further configured to perform any of the following:

[0162] The channel information is input into the first AI network model, and the output of the first AI network model is input into the second AI network model;

[0163] The channel information is input into the first AI network model, and the output of the target network structure in the first AI network model is input into the second AI network model.

[0164] Optionally, when the first channel feature information and the second channel feature information are reported via CSI, the reporting module 402 is further configured to perform any one of the following:

[0165] The first channel characteristic information is reported to the network-side device through the first CSI, and the second channel characteristic information is reported to the network-side device through the second CSI;

[0166] The first channel feature information and the second channel feature information are reported to the network-side device through the first CSI, wherein the first channel feature information is contained in the first part of the first CSI and the second channel feature information is contained in the second part of the first CSI.

[0167] Optionally, the first part may also include the length of the second channel feature information.

[0168] Optionally, the reporting module 402 includes:

[0169] The splicing unit is used to splice the first channel feature information and the second channel feature information to obtain the target channel feature information;

[0170] The reporting unit is used to report the target channel characteristic information to the network-side equipment.

[0171] Optionally, the splicing unit is used to perform any of the following:

[0172] The second channel feature information is concatenated after the first channel feature information to obtain the target channel feature information;

[0173] The first channel feature information and the second channel feature information are interleaved and spliced ​​together to obtain the target channel feature information;

[0174] The second channel feature information is concatenated after the first channel feature information, and the concatenated first channel feature information and second channel feature information are subjected to a first operation with a preset scrambling code sequence to obtain the target channel feature information;

[0175] The second channel feature information is combined with the first channel feature information to perform a second operation to obtain the third channel feature information. The third channel feature information is then concatenated with the first channel feature information to obtain the target channel feature information.

[0176] Optionally, the length of the first channel feature information is related to the preset scrambling code sequence.

[0177] Optionally, the reporting unit is further configured to:

[0178] The target channel feature information is quantized by a quantization network, and the quantized target channel feature information is reported to the network-side device.

[0179] Optionally, the acquisition module 401 is further configured to:

[0180] Obtain the first channel feature information output by the first AI network model before quantization and the second channel feature information output by the second AI network model before quantization;

[0181] The reporting module 402 is also used for:

[0182] The first channel feature information and the second channel feature information are concatenated to obtain the target channel feature information;

[0183] The target channel feature information is quantized by a quantization network, and the quantized target channel feature information is reported to the network-side device.

[0184] Optionally, the length of the first channel feature information is a fixed value, and the length of the second channel feature information is a fixed value or a variable value.

[0185] Optionally, the reporting module 402 is further configured to:

[0186] The first channel feature information is reported to the network-side device, and the second channel feature information is discarded.

[0187] Optionally, the device includes a first AI network model and at least one second AI network model, and the acquisition module 401 is further configured to:

[0188] Channel information is input into the first AI network model and the target second AI network model, wherein the at least one second AI network model includes the target second AI network model:

[0189] The target second AI network model is determined by at least one of the following:

[0190] Channel environment;

[0191] Instructions for the network-side devices.

[0192] Optionally, the first AI network model corresponds to the third AI network model of the network-side device, and the second AI network model corresponds to the fourth AI network model of the network-side device.

[0193] The input to the third AI network model is the first channel feature information, and the input to the fourth AI network model includes any one of the following:

[0194] The second channel feature information;

[0195] The first channel feature information and the second channel feature information;

[0196] The output of the first AI decoding network model and the second channel feature information.

[0197] Optionally, the second channel feature information includes N information blocks, which are arranged in a preset order, where N is an integer greater than 1;

[0198] When the fourth AI network model decodes the second channel feature information, the (i+1)th information block among the N information blocks is decoded based on the ith information block, where i is an integer less than N.

[0199] Optionally, if the second channel feature information is discarded, the N information blocks are discarded in reverse order of the preset order.

[0200] In this embodiment of the application, for channel information that is relatively long, the device and the network-side equipment can process the channel information through two sets of AI network models respectively. Each set of AI network models processes a portion of the channel information, thereby avoiding transmission errors and loss of channel information.

[0201] The channel feature information transmission device in this application embodiment can be an electronic device, such as an electronic device with an operating system, or a component in an electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal, or other devices besides a terminal. For example, the terminal can include, but is not limited to, the type of terminal 11 listed above; other devices can be servers, network attached storage (NAS), etc., and this application embodiment does not specifically limit the type.

[0202] The channel feature information transmission device provided in this application embodiment can achieve... Figure 2 The various processes implemented on the terminal in the method embodiment achieve the same technical effect, and will not be described again here to avoid repetition.

[0203] Please refer to Figure 5 , Figure 5 This is another channel feature information transmission device provided in the embodiments of this application, such as... Figure 5 As shown, the channel feature information transmission device 500 includes:

[0204] The receiving module 501 is used to receive at least one of the first channel characteristic information and the second channel characteristic information reported by the terminal;

[0205] Wherein, the first channel feature information is output by the terminal through the first AI network model, and the second channel feature information is output by the terminal through the second AI network model.

[0206] Optionally, when the first channel feature information and the second channel feature information are reported via CSI, the receiving module 501 is further configured to perform any one of the following:

[0207] The receiving terminal reports first channel characteristic information via the first CSI and second channel characteristic information via the second CSI;

[0208] The receiving terminal reports first channel characteristic information and second channel characteristic information through the first CSI, wherein the first channel characteristic information is contained in the first part of the first CSI and the second channel characteristic information is contained in the second part of the first CSI.

[0209] Optionally, the first part may also include the length of the second channel feature information.

[0210] Optionally, the receiving module 501 is further configured to:

[0211] The target channel feature information reported by the receiving terminal is channel feature information obtained by the terminal by concatenating the first channel feature information and the second channel feature information.

[0212] Optionally, the length of the first channel feature information is a fixed value, and the length of the second channel feature information is a fixed value or a variable value.

[0213] Optionally, the device includes a third AI decoding network model corresponding to the first AI network model and a fourth AI network model corresponding to the second AI network model;

[0214] The input to the third AI network model is the first channel feature information, and the input to the fourth AI network model includes any one of the following:

[0215] The second channel feature information;

[0216] The first channel feature information and the second channel feature information;

[0217] The output of the first AI decoding network model and the second channel feature information.

[0218] Optionally, the second channel feature information includes N information blocks, which are arranged in a preset order, where N is an integer greater than 1;

[0219] The device further includes a decoding module for:

[0220] The second channel feature information is decoded using the fourth AI network model, wherein the (i+1)th information block of the N information blocks is decoded based on the ith information block, where i is an integer less than N.

[0221] Optionally, the device further includes an update module for:

[0222] Update the first AI network model and the second AI network model; or,

[0223] The second AI network model is then updated.

[0224] Optionally, updating the second AI network model includes adjusting the length of the second channel feature information.

[0225] In this embodiment of the application, for channel information that is relatively long, the terminal and the device can process the channel information through two sets of AI network models respectively. Each set of AI network models processes a portion of the channel information, thereby avoiding transmission errors and loss of channel information.

[0226] The channel feature information transmission device provided in this application embodiment can achieve... Figure 3 The various processes implemented by the network-side device in the method embodiment achieve the same technical effect, and will not be described again here to avoid repetition.

[0227] Optional, such as Figure 6 As shown, this application embodiment also provides a communication device 600, including a processor 601 and a memory 602. The memory 602 stores a program or instructions that can run on the processor 601. For example, when the communication device 600 is a terminal, the program or instructions executed by the processor 601 implement the above-mentioned... Figure 2 Each step of the method embodiment described herein can achieve the same technical effect. When the communication device 600 is a network-side device, the program or instruction executed by the processor 601 implements the above-described steps. Figure 3 The steps of the method embodiments described herein can achieve the same technical effect, and will not be repeated here to avoid repetition.

[0228] This application embodiment also provides a terminal, including a processor and a communication interface. The processor is used to input channel information into a first artificial intelligence (AI) network model and a second AI network model, and to acquire first channel feature information output by the first AI network model and second channel feature information output by the second AI network model. The communication interface is used to report at least one of the first channel feature information and the second channel feature information to a network-side device. This terminal embodiment corresponds to the above-described terminal-side method embodiment. All implementation processes and methods of the above method embodiments can be applied to this terminal embodiment and achieve the same technical effect. Specifically, Figure 7 A schematic diagram of the hardware structure of a terminal to implement an embodiment of this application.

[0229] The terminal 700 includes, but is not limited to, at least some of the following components: radio frequency unit 701, network module 702, audio output unit 703, input unit 704, sensor 705, display unit 706, user input unit 707, interface unit 708, memory 709, and processor 710.

[0230] Those skilled in the art will understand that the terminal 700 may also include a power supply (such as a battery) for supplying power to various components. The power supply may be logically connected to the processor 710 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system. Figure 7 The terminal structure shown does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown, or combine certain components, or have different component arrangements, which will not be elaborated here.

[0231] It should be understood that, in this embodiment, the input unit 704 may include a graphics processing unit (GPU) 7041 and a microphone 7042. The GPU 7041 processes image data of still images or videos obtained by an image capture device (such as a camera) in video capture mode or image capture mode. The display unit 706 may include a display panel 7061, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, or the like. The user input unit 707 includes at least one of a touch panel 7071 and other input devices 7072. The touch panel 7071 is also called a touch screen. The touch panel 7071 may include a touch detection device and a touch controller. Other input devices 7072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, and joysticks, which will not be described in detail here.

[0232] In this embodiment, after receiving downlink data from the network-side device, the radio frequency unit 701 can transmit it to the processor 710 for processing; in addition, the radio frequency unit 701 can send uplink data to the network-side device. Typically, the radio frequency unit 701 includes, but is not limited to, an antenna, amplifier, transceiver, coupler, low-noise amplifier, duplexer, etc.

[0233] The memory 709 can be used to store software programs or instructions, as well as various data. The memory 709 may primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area may store the operating system, application programs or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, the memory 709 may include volatile memory or non-volatile memory, or both. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus RAM (DRRAM). The memory 709 in the embodiments of this application includes, but is not limited to, these and any other suitable types of memory.

[0234] Processor 710 may include one or more processing units; optionally, processor 710 integrates an application processor and a modem processor, wherein the application processor mainly handles operations involving the operating system, user interface, and applications, and the modem processor mainly handles wireless communication signals, such as a baseband processor. It is understood that the aforementioned modem processor may also not be integrated into processor 710.

[0235] The processor 710 is used to input channel information into the first artificial intelligence (AI) network model and the second AI network model, and to obtain the first channel feature information output by the first AI network model and the second channel feature information output by the second AI network model.

[0236] Radio frequency unit 701 is used to report at least one of the first channel characteristic information and the second channel characteristic information to the network side device.

[0237] Optionally, the processor 710 is further configured to perform at least one of the following:

[0238] The channel information is input into the first AI network model after undergoing the first preprocessing.

[0239] The channel information is then preprocessed and input into the second AI network model.

[0240] Optionally, the processor 710 is further configured to perform any of the following:

[0241] The channel information is input into the first AI network model, and the output of the first AI network model is input into the second AI network model;

[0242] The channel information is input into the first AI network model, and the output of the target network structure in the first AI network model is input into the second AI network model.

[0243] Optionally, when the first channel feature information and the second channel feature information are reported via CSI, the radio frequency unit 701 is further configured to perform any one of the following:

[0244] The first channel characteristic information is reported to the network-side device through the first CSI, and the second channel characteristic information is reported to the network-side device through the second CSI;

[0245] The first channel feature information and the second channel feature information are reported to the network-side device through the first CSI, wherein the first channel feature information is contained in the first part of the first CSI and the second channel feature information is contained in the second part of the first CSI.

[0246] Optionally, the first part may also include the length of the second channel feature information.

[0247] Optionally, the processor 710 is further configured to:

[0248] The first channel feature information and the second channel feature information are concatenated to obtain the target channel feature information;

[0249] The radio frequency unit 701 is also used to report the target channel characteristic information to the network-side device.

[0250] Optionally, the processor 710 is further configured to perform any of the following:

[0251] The second channel feature information is concatenated after the first channel feature information to obtain the target channel feature information;

[0252] The first channel feature information and the second channel feature information are interleaved and spliced ​​together to obtain the target channel feature information;

[0253] The second channel feature information is concatenated after the first channel feature information, and the concatenated first channel feature information and second channel feature information are subjected to a first operation with a preset scrambling code sequence to obtain the target channel feature information;

[0254] The second channel feature information is combined with the first channel feature information to perform a second operation to obtain the third channel feature information. The third channel feature information is then concatenated with the first channel feature information to obtain the target channel feature information.

[0255] Optionally, the length of the first channel feature information is related to the preset scrambling code sequence.

[0256] Optionally, the radio frequency unit 701 is further configured to:

[0257] The target channel feature information is quantized by a quantization network, and the quantized target channel feature information is reported to the network-side device.

[0258] Optionally, the processor 710 is further configured to:

[0259] Obtain the first channel feature information output by the first AI network model before quantization and the second channel feature information output by the second AI network model before quantization;

[0260] The first channel feature information and the second channel feature information are concatenated to obtain the target channel feature information;

[0261] The radio frequency unit 701 is also used for:

[0262] The target channel feature information is quantized by a quantization network, and the quantized target channel feature information is reported to the network-side device.

[0263] Optionally, the length of the first channel feature information is a fixed value, and the length of the second channel feature information is a fixed value or a variable value.

[0264] Optionally, the radio frequency unit 701 is further configured to:

[0265] The first channel feature information is reported to the network-side device, and the second channel feature information is discarded.

[0266] Optionally, the terminal includes a first AI network model and at least one second AI network model, and the processor 710 is further configured to:

[0267] Channel information is input into the first AI network model and the target second AI network model, wherein the at least one second AI network model includes the target second AI network model:

[0268] The target second AI network model is determined by at least one of the following:

[0269] Channel environment;

[0270] Instructions for the network-side devices.

[0271] Optionally, the first AI network model corresponds to the third AI network model of the network-side device, and the second AI network model corresponds to the fourth AI network model of the network-side device.

[0272] The input to the third AI network model is the first channel feature information, and the input to the fourth AI network model includes any one of the following:

[0273] The second channel feature information;

[0274] The first channel feature information and the second channel feature information;

[0275] The output of the first AI decoding network model and the second channel feature information.

[0276] Optionally, the second channel feature information includes N information blocks, which are arranged in a preset order, where N is an integer greater than 1;

[0277] When the fourth AI network model decodes the second channel feature information, the (i+1)th information block among the N information blocks is decoded based on the ith information block, where i is an integer less than N.

[0278] Optionally, if the second channel feature information is discarded, the N information blocks are discarded in reverse order of the preset order.

[0279] In this embodiment of the application, for long channel information, the terminal and the network-side device can process the channel information through two sets of AI network models respectively. Each set of AI network models processes a portion of the channel information, thereby avoiding transmission errors and loss of channel information.

[0280] This application also provides a network-side device, including a processor and a communication interface. The communication interface is used to receive at least one of a first channel feature information and a second channel feature information reported by a terminal. The first channel feature information is output by the terminal through a first AI network model, and the second channel feature information is output by the terminal through a second AI network model. This network-side device embodiment corresponds to the above-described network-side device method embodiment. All implementation processes and methods of the above method embodiments can be applied to this network-side device embodiment and achieve the same technical effects.

[0281] Specifically, embodiments of this application also provide a network-side device. For example... Figure 8 As shown, the network-side device 800 includes: an antenna 81, a radio frequency (RF) device 82, a baseband device 83, a processor 84, and a memory 85. The antenna 81 is connected to the RF device 82. In the uplink direction, the RF device 82 receives information through the antenna 81 and transmits the received information to the baseband device 83 for processing. In the downlink direction, the baseband device 83 processes the information to be transmitted and sends it to the RF device 82. The RF device 82 processes the received information and transmits it through the antenna 81.

[0282] The method executed by the network-side device in the above embodiments can be implemented in the baseband device 83, which includes a baseband processor.

[0283] Baseband device 83 may include, for example, at least one baseband board on which multiple chips are disposed, such as... Figure 8 As shown, one of the chips is, for example, a baseband processor, which is connected to the memory 85 via a bus interface to call the program in the memory 85 and execute the network device operation shown in the above method embodiment.

[0284] The network-side device may also include a network interface 86, such as a common public radio interface (CPRI).

[0285] Specifically, the network-side device 800 of this embodiment further includes: instructions or programs stored in a memory 85 and executable on a processor 84, wherein the processor 84 calls the instructions or programs in the memory 85 to execute. Figure 5 The methods executed by each module shown achieve the same technical effect, and to avoid repetition, they will not be described in detail here.

[0286] This application embodiment also provides a readable storage medium storing a program or instructions that, when executed by a processor, implement the above-described functionality. Figure 2 The various processes of the method embodiments, or the implementations of the above... Figure 3The various processes in the method embodiments described herein can achieve the same technical effect, and will not be repeated here to avoid repetition.

[0287] The processor is the processor in the terminal described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0288] This application embodiment also provides a chip, the chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the above. Figure 2 The various processes of the method embodiments, or the implementations of the above... Figure 3 The various processes in the method embodiments described herein can achieve the same technical effect, and will not be repeated here to avoid repetition.

[0289] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0290] This application embodiment also provides a computer program / program product, which is stored in a storage medium and executed by at least one processor to implement the above. Figure 2 The various processes of the method embodiments, or the implementations of the above... Figure 3 The various processes in the method embodiments described herein can achieve the same technical effect, and will not be repeated here to avoid repetition.

[0291] This application also provides a communication system, including: a terminal and a network-side device, wherein the terminal can be used to perform, for example... Figure 2 The network-side device can be used to execute the above-described steps of the channel feature information transmission method. Figure 3 The steps of the channel feature information transmission method described above.

[0292] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0293] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0294] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A channel characteristic information transmission method characterized by comprising: include: The terminal inputs channel information into the first AI network model and the second AI network model, and obtains the first channel feature information output by the first AI network model and the second channel feature information output by the second AI network model. The terminal reports at least one of the first channel feature information and the second channel feature information to the network-side device; When the first channel feature information and the second channel feature information are reported via Channel State Information (CSI), the terminal reports at least one of the first channel feature information and the second channel feature information to the network-side device, including any one of the following: The terminal reports the first channel feature information to the network-side device through the first CSI, and reports the second channel feature information to the network-side device through the second CSI; The terminal reports the first channel feature information and the second channel feature information to the network-side device through the first CSI, wherein the first channel feature information is contained in the first part of the first CSI and the second channel feature information is contained in the second part of the first CSI.

2. The method of claim 1, wherein, The terminal inputs channel information into the first AI network model and the second AI network model, including at least one of the following: The terminal inputs the channel information into the first AI network model after first preprocessing. The terminal inputs the channel information into the second AI network model after undergoing a second preprocessing step.

3. The method according to claim 2, characterized in that, The step of inputting the channel information into the second AI network model after second preprocessing includes any one of the following: The channel information is input into the first AI network model, and the output of the first AI network model is input into the second AI network model; The channel information is input into the first AI network model, and the output of the target network structure in the first AI network model is input into the second AI network model.

4. The method according to claim 1, characterized in that, The first part also includes the length of the second channel feature information.

5. The method according to claim 1, characterized in that, The terminal reports at least one of the first channel feature information and the second channel feature information to the network-side device, including: The terminal concatenates the first channel feature information and the second channel feature information to obtain the target channel feature information; The terminal reports the target channel characteristic information to the network-side device.

6. The method according to claim 5, characterized in that, The terminal concatenates the first channel feature information and the second channel feature information to obtain target channel feature information, including any one of the following: The terminal concatenates the second channel feature information after the first channel feature information to obtain the target channel feature information; The terminal interleaves and splices the first channel feature information and the second channel feature information to obtain the target channel feature information; The terminal concatenates the second channel feature information after the first channel feature information, and performs a first operation on the concatenated first channel feature information and second channel feature information with a preset scrambling code sequence to obtain target channel feature information; The terminal performs a second operation on the second channel feature information and the first channel feature information to obtain the third channel feature information, and then concatenates the third channel feature information after the first channel feature information to obtain the target channel feature information.

7. The method according to claim 6, characterized in that, The length of the first channel feature information is related to the preset scrambling code sequence.

8. The method according to claim 5, characterized in that, The terminal reports the target channel characteristic information to the network-side device, including: The terminal quantizes the target channel feature information through a quantization network and reports the quantized target channel feature information to the network-side device.

9. The method according to claim 1, characterized in that, The step of obtaining the first channel feature information output by the first AI network model and the second channel feature information output by the second AI network model includes: Obtain the first channel feature information output by the first AI network model before quantization and the second channel feature information output by the second AI network model before quantization; The terminal reports at least one of the first channel feature information and the second channel feature information to the network-side device, including: The terminal concatenates the first channel feature information and the second channel feature information to obtain the target channel feature information; The terminal quantizes the target channel feature information through a quantization network and reports the quantized target channel feature information to the network-side device.

10. The method according to any one of claims 1-9, characterized in that, The length of the first channel feature information is a fixed value, and the length of the second channel feature information is a fixed value or a variable value.

11. The method according to any one of claims 1-9, characterized in that, The terminal reports at least one of the first channel feature information and the second channel feature information to the network-side device, including: The terminal reports the first channel feature information to the network-side device and discards the second channel feature information.

12. The method according to any one of claims 1-9, characterized in that, The terminal includes a first AI network model and at least one second AI network model. The terminal inputs channel information into the first AI network model and the second AI network model, including: The terminal inputs channel information into the first AI network model and the target second AI network model, wherein the at least one second AI network model includes the target second AI network model: The target second AI network model is determined by at least one of the following: Channel environment; Instructions for the network-side devices.

13. The method according to any one of claims 1-9, characterized in that, The first AI network model corresponds to the third AI network model of the network-side device, and the second AI network model corresponds to the fourth AI network model of the network-side device. The input to the third AI network model is the first channel feature information, and the input to the fourth AI network model includes any one of the following: The second channel feature information; The first channel feature information and the second channel feature information; The output of the third AI network model and the second channel feature information.

14. The method according to claim 13, characterized in that, The second channel feature information includes N information blocks, which are arranged in a preset order, where N is an integer greater than 1; When the fourth AI network model decodes the second channel feature information, the (i+1)th information block among the N information blocks is decoded based on the ith information block, where i is an integer less than N.

15. The method according to claim 14, characterized in that, If the second channel feature information is discarded, the N information blocks are discarded in reverse order according to the preset order.

16. A method for transmitting channel feature information, characterized in that, include: The network-side device receives at least one of the first channel characteristic information and the second channel characteristic information reported by the terminal; Wherein, the first channel feature information is output by the terminal through the first AI network model, and the second channel feature information is output by the terminal through the second AI network model; When the first channel feature information and the second channel feature information are reported via Channel State Information (CSI), the network-side device receives at least one of the first channel feature information and the second channel feature information reported by the terminal, including any one of the following: The network-side device receives the first channel characteristic information reported by the terminal through the first CSI, and the second channel characteristic information reported through the second CSI; The network-side device receives first channel characteristic information and second channel characteristic information reported by the terminal through the first CSI, wherein the first channel characteristic information is contained in the first part of the first CSI and the second channel characteristic information is contained in the second part of the first CSI.

17. The method according to claim 16, characterized in that, The first part also includes the length of the second channel feature information.

18. The method according to claim 16, characterized in that, The network-side device receives at least one of the first channel characteristic information and the second channel characteristic information reported by the terminal, including: The network-side device receives target channel feature information reported by the terminal. The target channel feature information is channel feature information obtained by the terminal by concatenating the first channel feature information and the second channel feature information.

19. The method according to claim 16, characterized in that, The length of the first channel feature information is a fixed value, and the length of the second channel feature information is a fixed value or a variable value.

20. The method according to claim 16, characterized in that, The network-side device includes a third AI network model corresponding to the first AI network model and a fourth AI network model corresponding to the second AI network model. The input to the third AI network model is the first channel feature information, and the input to the fourth AI network model includes any one of the following: The second channel feature information; The first channel feature information and the second channel feature information; The output of the third AI network model and the second channel feature information.

21. The method according to claim 20, characterized in that, The second channel feature information includes N information blocks, which are arranged in a preset order, where N is an integer greater than 1; The method further includes: The network-side device decodes the second channel feature information through the fourth AI network model, wherein the (i+1)th information block among the N information blocks is decoded based on the ith information block, where i is an integer less than N.

22. The method according to claim 20, characterized in that, The method further includes: The network-side device updates the first AI network model and the second AI network model; or... The network-side device updates the second AI network model.

23. The method according to claim 22, characterized in that, The update of the second AI network model includes adjusting the length of the second channel feature information.

24. A channel feature information transmission device, characterized in that, include: The acquisition module is used to input channel information into the first artificial intelligence (AI) network model and the second AI network model, and to acquire the first channel feature information output by the first AI network model and the second channel feature information output by the second AI network model. The reporting module is used to report at least one of the first channel feature information and the second channel feature information to the network-side device; When the first channel feature information and the second channel feature information are reported via CSI, the reporting module is also configured to perform any one of the following: The first channel characteristic information is reported to the network-side device through the first CSI, and the second channel characteristic information is reported to the network-side device through the second CSI; The first channel feature information and the second channel feature information are reported to the network-side device through the first CSI, wherein the first channel feature information is contained in the first part of the first CSI and the second channel feature information is contained in the second part of the first CSI.

25. A channel feature information transmission device, characterized in that, include: A receiving module is configured to receive at least one of the first channel characteristic information and the second channel characteristic information reported by the terminal. Wherein, the first channel feature information is output by the terminal through the first AI network model, and the second channel feature information is output by the terminal through the second AI network model; When the first channel feature information and the second channel feature information are reported via Channel State Information (CSI), the receiving module is further configured to perform any one of the following: The receiving terminal reports first channel characteristic information via the first CSI and second channel characteristic information via the second CSI; The receiving terminal reports first channel characteristic information and second channel characteristic information through the first CSI, wherein the first channel characteristic information is contained in the first part of the first CSI and the second channel characteristic information is contained in the second part of the first CSI.

26. A terminal, characterized in that, It includes a processor and a memory, the memory storing programs or instructions that can run on the processor, the programs or instructions being executed by the processor to implement the steps of the channel feature information transmission method as described in any one of claims 1-15.

27. A network-side device, characterized in that, It includes a processor and a memory, the memory storing programs or instructions that can run on the processor, the programs or instructions being executed by the processor to implement the steps of the channel feature information transmission method as described in any one of claims 16-23.

28. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the channel feature information transmission method as described in any one of claims 1-15, or implement the steps of the channel feature information transmission method as described in any one of claims 16-23.

Images