Information processing method and apparatus, and communication device

Abstract

The present application belongs to the technical field of communications. Disclosed are an information processing method and apparatus, and a communication device. The information processing method of the embodiments of the present application comprises: a communication device uses artificial intelligence (AI) units corresponding to N ranks to respectively process channel information of all layers of the N ranks, wherein the AI units corresponding to each rank comprise AI units corresponding to all layers of each rank, the N ranks comprise at least one rank greater than 1, and N is a positive integer.

Description

Information processing method, device and communication equipment

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to Chinese Patent Application No. 202410015948.8 filed on January 4, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present application belongs to the field of communication technology, and specifically relates to an information processing method, apparatus and communication equipment. Background Art

[0004] Currently, in mobile communication networks, tasks can be performed or services can be provided based on artificial intelligence (AI). For example, a terminal can encode channel state information (CSI) based on a pre-trained AI unit or AI model, and a network-side device can decode CSI based on a pre-trained AI unit or AI model. However, in the prior art, single-layer (i.e., rank 1) channel information is processed based on AI units or AI models. When there is a rank greater than 1, there is no relevant solution for how to process the channel information of a rank greater than 1 based on the AI ​​unit or AI model, which in turn affects the effect of AI-based processing of channel information of a rank greater than 1. Summary of the Invention

[0005] The embodiments of the present application provide an information processing method, apparatus, and communication equipment, which can provide a method for processing channel information of a rank greater than 1 based on an AI unit when there is a rank greater than 1, that is, combining the AI ​​units corresponding to all layers of the rank greater than 1 to obtain the AI ​​unit corresponding to the rank greater than 1 to process the channel information of each layer of the rank greater than 1.

[0006] In a first aspect, an information processing method is provided, the method comprising:

[0007] The communication device uses artificial intelligence AI units corresponding to N ranks to process channel information of each layer of the N ranks respectively;

[0008] The AI ​​unit corresponding to each of the ranks includes the AI ​​units corresponding to all layers of each of the ranks, the N ranks include at least one rank greater than 1, and N is a positive integer.

[0009] In a second aspect, an information processing device is provided, the device comprising:

[0010] A processing module, configured to use artificial intelligence AI units corresponding to N ranks to process channel information of each layer of the N ranks respectively;

[0011] The AI ​​unit corresponding to each of the ranks includes the AI ​​units corresponding to all layers of each of the ranks, the N ranks include at least one rank greater than 1, and N is a positive integer.

[0012] In a third aspect, a communication device is provided, which includes a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the program or instructions are executed by the processor, the steps of the method described in the first aspect are implemented.

[0013] In a fourth aspect, a communication device is provided, comprising a processor and a communication interface, wherein the processor is used to use artificial intelligence AI units corresponding to N ranks to process the channel information of each layer of the N ranks respectively; wherein the AI ​​units corresponding to each of the ranks respectively include AI units corresponding to all layers of each of the ranks, the N ranks include at least one rank greater than 1, and N is a positive integer.

[0014] In a fifth aspect, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0015] In a sixth aspect, a chip is provided, comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is configured to run a program or instruction to implement the steps of the method described in the first aspect.

[0016] In a seventh aspect, a computer program / program product is provided, wherein the computer program / program product is stored in a storage medium and is executed by at least one processor to implement the steps of the method described in the first aspect.

[0017] In an embodiment of the present application, the communication device uses artificial intelligence AI units corresponding to N ranks to process the channel information of each layer of the N ranks respectively; wherein, the AI ​​units corresponding to each of the ranks respectively include AI units corresponding to all layers of each of the ranks, and the N ranks include at least one rank greater than 1, and N is a positive integer. That is, the embodiment of the present application provides a method for processing channel information of a rank greater than 1 based on an AI unit when there is a rank greater than 1, that is, combining the AI ​​units corresponding to all layers of the rank greater than 1 to obtain the AI ​​unit corresponding to the rank greater than 1 to process the channel information of each layer of the rank greater than 1, which is conducive to improving the effect of processing channel information of a rank greater than 1 based on AI. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] FIG1 is a block diagram of a wireless communication system applicable to embodiments of the present application;

[0019] FIG2a is a schematic diagram of the structure of a neural network provided in an embodiment of the present application;

[0020] FIG2 b is a schematic diagram of the structure of a neuron provided in an embodiment of the present application;

[0021] FIG2c is a schematic diagram of time-frequency-spatial domain CSI encoding and decoding provided by an embodiment of the present application;

[0022] FIG3 is a flow chart of an information processing method provided in an embodiment of the present application;

[0023] 4a to 41 are schematic diagrams of intermediate information transmission provided by an embodiment of the present application;

[0024] 5a to 5j are schematic diagrams of intermediate information transmission provided by another embodiment of the present application;

[0025] FIG6 is a structural diagram of an information processing device provided in an embodiment of the present application;

[0026] FIG7 is a structural diagram of a communication device provided in an embodiment of the present application;

[0027] FIG8 is a structural diagram of a terminal provided in an embodiment of the present application;

[0028] FIG9 is a structural diagram of a network-side device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0029] The following will be combined with the accompanying drawings in the embodiments of this application to clearly describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of this application.

[0030] The terms "first", "second", etc. in this application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable where appropriate, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same type, and do not limit the number of objects, for example, the first object can be one or more. In addition, "or" in this application represents at least one of the connected objects. For example, "A or B" covers three options, namely, Option 1: including A but not including B; Option 2: including B but not including A; Option 3: including both A and B. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.

[0031] The term "indication" in this application can be either a direct indication (or explicit indication) or an indirect indication (or implicit indication). A direct indication can be understood as the sender explicitly informing the receiver of specific information, the operation to be performed, or the requested result, etc. in the instruction sent; an indirect indication can be understood as the receiver determining the corresponding information based on the instruction sent by the sender, or making a judgment and determining the operation to be performed or the requested result, etc. based on the judgment result.

[0032] It is worth noting that the technology described in the embodiments of the present application is not limited to the Long Term Evolution (LTE) / LTE-Advanced (LTE-A) system, 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) or other systems. The terms "system" and "network" in the embodiments of the present application are often used interchangeably, and the technology described can be used for the systems and radio technologies mentioned above, as well as for other systems and radio technologies. The following description describes a New Radio (NR) system for illustrative purposes, and NR terminology is used in most of the following description, but these technologies can also be applied to systems other than NR systems, such as 6th generation (6G) systems. th Generation, 6G) communication system.

[0033] FIG1 is a block diagram of a wireless communication system applicable to an embodiment of the present application. The wireless communication system includes a terminal 11 and a network-side device 12. The terminal 11 may be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer), a notebook computer, a personal digital assistant (PDA), a handheld computer, a netbook, an ultra-mobile personal computer (UMPC), a mobile internet device (MID), an augmented reality (AR), a virtual reality (VR) device, a robot, a wearable device (Wearable Device), an aircraft (Flight Vehicle), a vehicle-mounted device (VUE), a ship-mounted device, a pedestrian user equipment (PUE), a smart home (home appliances with wireless communication capabilities, such as refrigerators, televisions, washing machines, or furniture), a game console, a personal computer (PC), an ATM, or a self-service machine, or other terminal-side devices. Wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. Among them, the vehicle-mounted device can also be called a vehicle-mounted terminal, a vehicle-mounted controller, a vehicle-mounted module, a vehicle-mounted component, a vehicle-mounted chip or a vehicle-mounted unit, etc. It should be noted that the specific type of the terminal 11 is not limited in the embodiment of the present application. The network side device 12 may include an access network device or a core network device, wherein the access network device may also be called a radio access network (Radio Access Network, RAN) device, a radio access network function or a radio access network unit. The access network device may include a base station, a wireless local area network (WLAN) access point (AP) or a wireless fidelity (WiFi) node, etc.Among them, the base station can be referred to as Node B (NB), Evolved Node B (eNB), the next generation Node B (gNB), New Radio Node B (NR Node B), access point, Relay Base Station (RBS), Serving Base Station (SBS), Base Transceiver Station (BTS), radio base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), Home Node B (HNB), Home Evolved Node B (home evolved Node B), Transmission Reception Point (TRP) or other appropriate terms in the field. As long as the same technical effect is achieved, the base station is not limited to specific technical vocabulary. It should be noted that in the embodiment of the present application, only the base station in the NR system is used as an example for introduction, and the specific type of the base station is not limited.

[0034] The core network equipment may include but is not limited to at least one of the following: core network node, core network function, mobility management entity (MME), access mobility management function (AMF), session management function (SMF), user plane function (UPF), policy control function (PCF), policy and charging rules function unit (PCRF), edge application service discovery function (EASDF), unified data management (UDM), unified data repository (UDR), home user server (HSS), centralized network configuration (CNC), network storage function (NRF), network exposure function (NEF), local NEF (L-NEF), binding support function (BSF), application function ( It should be noted that in the embodiments of the present application, only the core network device in the NR system is introduced as an example, and the specific type of the core network device is not limited.

[0035] For ease of understanding, some of the contents involved in the embodiments of this application are described below:

[0036] 1. Artificial Intelligence (AI)

[0037] Artificial intelligence (AI) is currently being widely applied in various fields. Integrating AI into wireless communication networks to significantly improve technical indicators such as throughput, latency, and user capacity is a key task for future wireless communication networks. AI modules can be implemented in a variety of ways, such as neural networks, decision trees, support vector machines, and Bayesian classifiers. This application uses neural networks as an example, but does not limit the specific type of AI module.

[0038] For example, a neural network can be shown in Figure 2a. The neural network is composed of neurons, and each neuron can be shown in Figure 2b. Here, a1, a2, ... aK are inputs, w is the weight (multiplicative coefficient), b is the bias (additive coefficient), and σ(.) is the activation function. Common activation functions include Sigmoid, tanh, and Rectified Linear Unit (ReLU).

[0039] Neural network parameters are optimized using a gradient optimization algorithm. Gradient optimization algorithms are a class of algorithms that minimize or maximize an objective function (also called a loss function), which is often a mathematical combination of model parameters and data. For example, given data X and its corresponding label Y, we construct a neural network model f(.). With this model, we can obtain a predicted output f(x) based on the input x, and calculate the difference between the predicted value and the true value (f(x) - Y). This is the loss function. The goal is to find the appropriate W, b to minimize the value of this loss function. The smaller the loss value, the closer the model is to the true value.

[0040] Currently, most common optimization algorithms are based on the back propagation (BP) algorithm. The basic idea of ​​the BP algorithm is that the learning process consists of two steps: forward propagation of signals and back propagation of errors. During forward propagation, input samples are passed from the input layer, processed layer by layer through each hidden layer, and then transmitted to the output layer. If the actual output of the output layer does not match the expected output, the error begins back propagation. Back propagation involves propagating the output error back through the hidden layers to the input layer layer by layer in some form, distributing the error to all units in each layer. This error signal is then generated for each unit in each layer, which serves as the basis for adjusting the weights of each unit. This process of adjusting the weights of each layer, through forward propagation of signals and back propagation of errors, is repeated over and over again. This process of continuous weight adjustment is the network's learning and training process. This process continues until the error in the network output is reduced to an acceptable level, or until a pre-set number of learning cycles is reached.

[0041] Common optimization algorithms include gradient descent, stochastic gradient descent (SGD), mini-batch gradient descent, momentum method (Momentum), Nesterov (the name of the inventor, specifically stochastic gradient descent with momentum), adaptive gradient descent (Adagrad), Adadelta, root mean square error prop (RMSprop), adaptive momentum estimation (Adam), etc.

[0042] When these optimization algorithms backpropagate errors, they all calculate the derivative / partial derivative of the current neuron based on the error / loss obtained by the loss function, add the influence of the learning rate, the previous gradient / derivative / partial derivative, etc., obtain the gradient, and pass the gradient to the previous layer.

[0043] 2. AI Unit / AI Model

[0044] The AI ​​unit / AI model of the embodiment of the present application may also be referred to as a machine learning (ML) model, ML unit, AI structure, AI function, AI feature, machine learning model, neural network, neural network function, neural network function, etc., or the above-mentioned AI unit / AI model may also refer to a processing unit that can implement specific algorithms, formulas, processing procedures, capabilities, etc. related to AI, or the AI ​​unit / AI model may be a processing method, algorithm, function, module or unit for a specific data set, or the AI ​​unit / AI model may be a processing method, algorithm, function, module or unit running on AI / ML related hardware such as a graphics processing unit (GPU), a neural network processor (NPU), a tensor processing unit (TPU) or an application-specific integrated circuit (ASIC), and the embodiment of the present application does not specifically limit this. Optionally, the specific data set includes at least one of the input and output of the AI ​​unit / AI model.

[0045] Optionally, the identifier of the AI ​​unit / AI model may be an AI model identifier, an AI structure identifier, an AI algorithm identifier, or an identifier of a specific data set associated with the AI ​​unit / AI model, or an identifier of a specific scenario, environment, channel feature, or device related to the AI / ML, or an identifier of a function, feature, capability, or module related to the AI / ML. This embodiment of the present application does not specifically limit this.

[0046] 3. Channel State Information (CSI) Compression in Time-Frequency-Space Domain

[0047] For example, as shown in Figure 2c, in traditional space-frequency domain compression, the CSI information (space-frequency domain channel) of time slot (Slot) X is input into encoder (Encoder, ENC) X to generate CSI reporting information (feedback information (Feedback) in Figure 2c); the network (Network, NW) receives the CSI reporting information and decodes it through decoder (Decoder, DEC) X to obtain the recovered CSI information (i.e., CSI').

[0048] In space-time-frequency domain compression, ENC X uses the internal information of ENC X-1, that is, uses previous historical information to assist in encoding. This information is sent from ENC X (i.e., the encoder of slot X) to ENC X+1 (the encoder of slot X+1). Optionally, Slot X may refer to the time slot or time at which the CSI information input to the encoder is currently obtained, the time slot or time at which the encoder is currently encoding, the time slot or time at which CSI reporting is currently performed, the time slot or time at which the decoder is currently obtaining CSI reporting information, the time slot or time at which the decoder is currently performing inference, or the time slot or time at which the decoder is currently obtaining recovered CSI information. Slot X+1 may refer to the time slot or time at which the CSI information input to the encoder is next obtained, the time slot or time at which the encoder is currently encoding, the time slot or time at which CSI reporting is next performed, the time slot or time at which the decoder is next obtaining CSI reporting information, the time slot or time at which the decoder is next performing inference, or the time slot or time at which the decoder is next obtaining recovered CSI information. Optionally, to account for the impact of hardware processing delay, Slot X or Slot X+1 may also be the time slot or time associated with the encoder or decoder minus a specific time advance or plus a specific delay.

[0049] It should be noted that the intermediate information of the decoder is the same as the intermediate information of the encoder, and will not be described in detail here. In addition, the intermediate state information of the encoder is generally only transmitted between encoders; the intermediate state information of the decoder is generally only transmitted between decoders.

[0050] The information processing method provided by the embodiments of the present application is described in detail below through some embodiments and their application scenarios in conjunction with the accompanying drawings.

[0051] Please refer to Figure 3, which is a flowchart of an information processing method provided in an embodiment of the present application. The method can be executed by a communication device, wherein the communication device can be a terminal or a network side device, etc.

[0052] As shown in FIG3 , the information processing method provided in the embodiment of the present application includes the following steps:

[0053] Step 301: The communication device uses AI units corresponding to N ranks to process channel information of each layer of the N ranks respectively;

[0054] The AI ​​unit corresponding to each of the ranks includes the AI ​​units corresponding to all layers of each of the ranks, the N ranks include at least one rank greater than 1, and N is a positive integer.

[0055] In this embodiment, the AI ​​units described above can be referred to in the aforementioned related descriptions and will not be described in detail here. For example, the AI ​​units corresponding to the ranks described above can also be referred to as the AI ​​models corresponding to the ranks; the AI ​​units corresponding to the layers described above can also be referred to as the AI ​​models or AI sub-models corresponding to the layers.

[0056] The AI ​​units corresponding to each of the above ranks respectively include the AI ​​units corresponding to all layers of each rank. For example, if the above N ranks include the first rank and the second rank, the AI ​​units corresponding to the first rank include the AI ​​units corresponding to all layers of the first rank, and the AI ​​units corresponding to the second rank include the AI ​​units corresponding to all layers of the second rank; if the above N ranks include only one rank, for example, Rank 4 (i.e., rank 4), the AI ​​units corresponding to Rank 4 include the AI ​​units corresponding to all layers of Rank 4 (i.e., Layer 0 to Layer 3).

[0057] It is understood that when N is 1, the communication device uses the AI ​​unit corresponding to one rank to process the channel information of each layer of that rank; when N is greater than 1, the communication device uses the AI ​​unit corresponding to each rank to process the channel information of each layer of each rank separately. When N is greater than 1, the values ​​of the N ranks can be the same, for example, the N ranks include Rank 4 corresponding to N different time units; or they can be different, for example, the N ranks include Rank 1, Rank 2, and Rank 4.

[0058] The AI ​​units corresponding to the N ranks respectively process the channel information of each layer of the N ranks, that is, the AI ​​unit corresponding to each rank is used to process the channel information of each layer of each rank. For example, if the N ranks include a first rank and a second rank, the AI ​​unit corresponding to the first rank is used to process the channel information of each layer of the first rank, and the AI ​​unit corresponding to the second rank is used to process the channel information of each layer of the second rank.

[0059] Among them, the channel information of the above-mentioned layers may include but is not limited to channel information collected from various reference signals or channels; among them, the reference signal may include but is not limited to synchronization signal block (Synchronization Signal and PBCH block, SSB), channel state information reference signal (CSI-RS), tracking reference signal (TRS), phase tracking reference signal (PTRS), sounding reference signal (SRS), etc.; the channel may include but is not limited to physical downlink shared channel (PDSCH), physical uplink shared channel (PUSCH), physical downlink control channel (PDCCH), physical uplink control channel (PUCCH), etc. Among them, the channel information may include but is not limited to frequency domain channel information, time domain channel information, Doppler domain channel information, delay domain channel information, etc.; optionally, the time domain channel information includes multipath power information, multipath delay information or multipath phase information of the time domain channel impulse response, such as power delay profile (PDP), multipath delay profile (DP) or time domain channel impulse response (CIR).

[0060] The AI ​​units corresponding to different layers of each rank mentioned above may be the same or different. It is understandable that if the AI ​​units corresponding to different layers of the same rank are the same AI unit, then the AI ​​unit corresponding to the rank includes one AI unit. In addition, the AI ​​unit corresponding to each layer is processed based on the channel information corresponding to each layer, that is, the channel information of each layer is used as the input information of the AI ​​unit corresponding to each layer. In some examples, if there are multiple layers corresponding to the same AI unit, the AI ​​unit can be used to process based on the channel information corresponding to each of the multiple layers mentioned above in sequence.

[0061] Exemplarily, if the communication device is a terminal, the AI ​​unit may be an AI unit for encoding or compressing channel information, and the channel information of each layer of the N ranks is the channel information before compression or encoding; if the communication device is a network-side device, the AI ​​unit may be an AI unit for decoding or decompressing channel information, and the channel information of each layer of the N ranks is the channel information after compression or encoding. The following examples illustrate the embodiments of the present application in different situations:

[0062] Case 1: The communication device is a terminal.

[0063] The terminal uses AI units corresponding to N ranks (for example, encoding or compression AI units) to encode or compress the CSI information of each layer of the N ranks respectively, and obtains CSI reporting information, that is, the channel information of each layer of the above N ranks after encoding or compression processing, and can send the CSI reporting information to the network side device.

[0064] Case 2: The communication device is a network-side device.

[0065] The network-side device receives CSI reporting information from the terminal, which includes CSI information of each layer of N ranks, wherein the channel information is CSI information after encoding or compression processing, and uses AI units corresponding to the N ranks (for example, decoding or decompression AI units) to decode or decompress the channel information of each layer of the N ranks respectively to obtain the recovered CSI information of each layer of the N ranks.

[0066] In an embodiment of the present application, the communication device uses artificial intelligence AI units corresponding to N ranks to process the channel information of each layer of the N ranks respectively; wherein, the AI ​​unit corresponding to each of the ranks respectively includes the AI ​​units corresponding to all layers of each of the ranks, and the N ranks include at least one rank greater than 1, and N is a positive integer. That is, the embodiment of the present application provides a method for processing the channel information of a rank greater than 1 based on an AI unit when there is a rank greater than 1, that is, the AI ​​units corresponding to all layers of the rank greater than 1 are combined to obtain the AI ​​unit corresponding to the rank greater than 1 to process the channel information of each layer of the rank greater than 1, which is conducive to improving the effect of processing the channel information of a rank greater than 1 based on AI.

[0067] Optionally, different ranks among the N ranks correspond to different AI units, and the same rank among the N ranks corresponds to the same AI unit.

[0068] In this embodiment, the same rank corresponds to the same AI unit, and different ranks correspond to different AI units. For example, rank 1, rank 2, rank 3, and rank 4 correspond to different AI units. For the AI ​​unit corresponding to each rank, its input includes the channel information of all layers of the rank, wherein the channel information of each layer of the rank is respectively used as the input information of the AI ​​unit corresponding to the layer in the AI ​​unit corresponding to the rank. For example, the channel information of Layer 0 of rank 2 is used as the input information of the AI ​​unit corresponding to Layer 0 in the AI ​​unit corresponding to rank 2, and the channel information of Layer 1 of rank 2 is used as the input information of the AI ​​unit corresponding to Layer 1 in the AI ​​unit corresponding to rank 2.

[0069] It should be noted that different ranks among the above N ranks can be understood as ranks with different values. The same rank among the above N ranks can be understood as ranks with the same value, for example, rank 2 corresponding to different time units.

[0070] It should also be noted that in this embodiment, different layers of the same rank may correspond to the same AI unit or may correspond to different AI units. The same layers of different ranks may correspond to different AI units. For example, the AI ​​unit corresponding to Layer 0 of Rank 2 is different from the AI ​​unit corresponding to Layer 0 of Rank 3. In some examples, when different layers of the same rank correspond to different AI units, the same layers of different ranks may correspond to the same AI unit.

[0071] In this embodiment, the AI ​​units corresponding to different ranks are different, and the AI ​​units corresponding to the same rank are the same, that is, the AI ​​units are divided from the rank dimension, so that different ranks use different AI units, reducing the difficulty of rank generalization of the AI ​​units, which makes it easier to process channel information of different ranks based on the AI ​​units.

[0072] Optionally, different layers of each rank correspond to different AI units.

[0073] In this embodiment, different layers of the same rank correspond to different AI units. For example, the AI ​​unit corresponding to Layer 0 of Rank 2 is different from the AI ​​unit corresponding to Layer 1.

[0074] Optionally, the same layer in different ranks among the N ranks corresponds to the same AI unit, and different layers of the same rank among the N ranks correspond to different AI units.

[0075] Exemplarily, if N ranks include Rank 2, Rank 3, and Rank 4, the same layers of Rank 2, Rank 3, and Rank 4 correspond to the same AI unit. For example, Layer 0 of Rank 2, Layer 0 of Rank 3, and Layer 0 of Rank 4 all correspond to the first AI unit, Layer 1 of Rank 2, Layer 1 of Rank 3, and Layer 1 of Rank 4 all correspond to the second AI unit, Layer 2 of Rank 3 and Layer 2 of Rank 4 all correspond to the third AI unit, and Layer 3 of Rank 4 corresponds to the fourth AI unit. The above-mentioned first to fourth AI units are different AI units.

[0076] It can be understood that the input information of the AI ​​unit corresponding to each layer includes the channel information of each layer.

[0077] In this embodiment, the same AI unit corresponds to the same layer in different ranks, and different AI units correspond to different layers of the same rank. This means that AI units are divided based on the layer dimension, leveraging the different distributions of channel features across different layers to enable AI units to learn the characteristics of specific layers, reducing the difficulty of layer generalization for the AI ​​units. This facilitates processing of channel information from different layers based on the AI ​​units. Furthermore, the same AI unit corresponds to the same layer in different ranks, which reduces the difficulty of rank generalization for the AI ​​units.

[0078] Optionally, the same layer in different ranks among the N ranks corresponds to the same AI unit, and different layers of the same rank among the N ranks correspond to the same AI unit.

[0079] In this embodiment, different ranks and different layers correspond to the same AI unit. For example, Layer 0, Layer 1, Layer 2, and Layer 3 of Rank 4 all correspond to the first AI unit, and Layer 0 of Rank 2 and Layer 2 of Rank 3 also correspond to the first AI unit.

[0080] It should be noted that the input information of the above-mentioned AI unit includes the channel information of one layer.

[0081] In this embodiment, the same AI unit corresponds to the same layer in different ranks, and different layers in the same rank correspond to the same AI unit. In other words, all layers of all ranks share a single AI unit. This reduces the number of AI units and facilitates their management and alignment. However, in this case, the AI ​​unit faces both rank generalization and layer generalization issues.

[0082] The above different implementations require a trade-off between generalization and the number of AI units. For scenarios with complex channel characteristics and no limit on the number of AI units, different ranks can be assigned different AI units. For scenarios where channel generalization is relatively easy to resolve and the number of AI units is limited, all layers of all ranks can share a single AI unit. For scenarios where generalization is easy to resolve for different ranks but difficult to resolve for different layers, the same AI unit can be assigned to the same layers of different ranks.

[0083] Optionally, the AI ​​unit corresponding to the first target layer of the first rank of the N ranks performs processing based on the first information corresponding to the first target layer;

[0084] Among them, the first information corresponding to the first target layer includes channel information of the first target layer and intermediate information of the associated layer of the first target layer, the associated layer of the first target layer includes at least one layer of all layers of the N ranks except the first target layer, and the intermediate information of the associated layer of the first target layer is the output information of the AI ​​unit corresponding to the associated layer of the first target layer.

[0085] In this embodiment, the first rank may include any rank or a specific rank among the N ranks, and the first target layer may include any layer or a specific layer of the first rank. The associated layer of the first target layer may include at least one layer of all layers of the N ranks except the first target layer. Exemplarily, the associated layer of the first target layer may include at least one layer of the first rank except the first target layer, or may include at least one layer of all layers of the other ranks of the N ranks except the first rank.

[0086] The intermediate information of the associated layer of the above-mentioned first target layer is part of the output information of the AI ​​unit corresponding to the associated layer of the first target layer, that is, part of the output information obtained by the AI ​​unit corresponding to the associated layer of the first target layer according to the channel information of the associated layer of the first target layer. Exemplarily, for a CSI encoder, its output information includes intermediate information and CSI reporting information; for a CSI decoder, its output information includes intermediate information and recovered channel information. In traditional space-frequency domain CSI compression, the output information of the CSI encoder is only CSI reporting information, and the output information of the CSI decoder is only recovered channel information. It should be noted that the AI ​​unit corresponding to the associated layer of the above-mentioned first target layer can only process the channel information of the associated layer of the first target layer, or can process the channel information of the associated layer of the first target layer and other input information (for example, the intermediate information of the associated layer of the associated layer of the first target layer).

[0087] It should also be noted that, generally speaking, the above-mentioned intermediate information will not be used as the output of the AI ​​unit (e.g., CSI module / CSI encoder / CSI decoder) on which it acts and passed to the next processing module. For example, the intermediate information output by the UE's CSI encoder will not be used as the UE's CSI reporting information to be reported to the network (NW); and the intermediate information output by the NW's decoder will not be used as the channel information recovered by the NW and passed to the next processing module of the NW.

[0088] In an embodiment of the present application, the AI ​​unit corresponding to the first target layer of the first rank among N ranks is processed based on the channel information of the first target layer and the intermediate information of the associated layer of the first target layer. This is conducive to improving the processing effect of the channel information of the first target layer based on the AI ​​unit. For example, if the above-mentioned AI unit is an AI compression unit, the AI ​​compression unit is compressed based on the channel information of the first target layer and the intermediate information of the associated layer of the first target layer, which is conducive to improving the compression ratio of the channel information of the first target layer.

[0089] Optionally, the intermediate information of the associated layer of the first target layer is the input information of the input layer of the AI ​​unit corresponding to the first target layer, or the intermediate information of the associated layer of the first target layer is the input information of at least one hidden layer of the AI ​​unit corresponding to the first target layer.

[0090] In this embodiment, the intermediate information of the associated layer of the first target layer can be input into the input layer (Input Layer) of the AI ​​unit corresponding to the first target layer; or the intermediate information of the associated layer of the first target layer can be input into at least one hidden layer (Hidden Layer) of the AI ​​unit corresponding to the first target layer, that is, the intermediate information of the associated layer of the first target layer acts on the intermediate layer of the AI ​​unit.

[0091] Optionally, the AI ​​unit corresponding to the second target layer of the second rank among the N ranks performs processing based on the second information corresponding to the second target layer;

[0092] The second rank is the rank corresponding to the first time unit among the N ranks, the second information corresponding to the second target layer includes channel information of the second target layer and intermediate information of an associated layer of the second target layer, and the intermediate information of the associated layer of the second target layer is output information of the AI ​​unit corresponding to the associated layer of the second target layer;

[0093] The associated layers of the second target layer include at least one of the following: a first layer, a second layer, and a third layer;

[0094] The first layer is a layer of the second rank, and the first layer is one layer lower than the second target layer, or the index of the first layer is the index of the second target layer minus 1;

[0095] The second layer is a layer of a third rank, the third rank being a rank among the N ranks corresponding to a second time unit, the second time unit being a time unit before the first time unit, and an index of the second layer is the same as an index of the second target layer;

[0096] The third layer is the topmost layer or the layer with the largest index among all layers of the fourth rank, or the third layer is the topmost layer or the layer with the largest index among all layers whose corresponding AI unit in the layer of the fourth rank uses the intermediate information of the lower layer for processing, and the fourth rank is the rank corresponding to the third time unit among the N ranks, and the third time unit is the time unit before the first time unit.

[0097] In this embodiment, the second rank may include any rank or a specific rank of the N ranks, and the second target layer may include any layer or a specific layer of the second rank.

[0098] The first time unit may be any time unit, and the time unit may include but is not limited to a time slot, a sub-time slot, a frame or a sub-frame.

[0099] The second time unit is a time unit before the first time unit. Exemplarily, the second time unit may be the K1 time units before the first time unit, where K1 is a positive integer. For example, the value of K1 may be 1 or 2. In some optional embodiments, the second time unit is a time unit in which the AI ​​unit was most recently used to process channel information before the first time unit, that is, the time unit in which the AI ​​unit was last used, or a time unit in which the same channel information was measured that was closest to the time unit in which the current channel information was measured.

[0100] The third time unit and the second time unit may be the same time unit or different time units. Exemplarily, the second time unit is the K2 time units before the first time unit, where K2 is a positive integer, for example, the value of K2 may be 1 or 2. In some optional embodiments, the third time unit may be the time unit in which the AI ​​unit was most recently used to process the channel information before the first time unit, that is, the time unit in which the AI ​​unit was last used, or the time unit in which the same channel information is measured that is closest to the time unit in which the current channel information is measured.

[0101] It can be understood that, when the second time unit and the third time unit are the same time unit, the third rank and the fourth rank are the same rank.

[0102] It should be noted that, when the second rank includes multiple ranks, the types of associated layers of the second target layers of different ranks in the multiple ranks may be the same, or may be different, wherein the types of associated layers include the first layer, the second layer, and the third layer. For example, the associated layer of the second target layers of some ranks is the first layer, the associated layer of the second target layers of some ranks is the third layer, the associated layers of the second target layers of some ranks include the first layer and the second layer, etc.

[0103] It should also be noted that, when the second target layer includes multiple layers, the AI ​​unit corresponding to each of the multiple layers performs processing based on the second information corresponding to the layer. Furthermore, the types of associated layers for different layers in the multiple layers may be different, for example, the associated layer for some layers may be the first layer, the associated layer for some layers may be the second layer, or the associated layers for some layers may include the first layer and the second layer, etc.; or the types of associated layers for different layers in the multiple layers may be the same, for example, the associated layer for each of the multiple layers may be the first layer, or the second layer, or the first layer and the second layer, etc. For example, the second target layer includes layers 1, 2, and 3 of the first time unit, wherein the associated layer of layer 1 of the first time unit may include the first layer, i.e., the associated layer of layer 1 of the first time unit is layer 0 of the first time unit; the associated layer of layer 2 of the first time unit may include the first layer and the second layer, i.e., the associated layer of layer 2 of the first time unit may include the first layer and the second layer, i.e., the associated layer of layer 2 of the first time unit may include the second layer, i.e., the associated layer of layer 3 of the first time unit may be layer 3 of the second time unit.

[0104] For example, the intermediate information of the first layer can be understood as the intermediate information transmitted in a first transmission mode, wherein the first transmission mode is: transmitting the intermediate information of Layer x at time t to Layer (x+1) at time t, x <K t -1,K t Indicates the value of the rank corresponding to time t.

[0105] The intermediate information of the second layer can be understood as the intermediate information transmitted in the second transmission mode, wherein the second transmission mode is: transmitting the intermediate information of Layer x at time t to Layer x at time t+1 or Layer x at the time closest to time t. <K t -1 and x <K t+1 -1, the above K t+1 Indicates the value of the rank corresponding to time t+1.

[0106] The intermediate information of the third layer can be understood as the intermediate information transmitted in accordance with the third transmission mode, wherein the third transmission mode is: Layer (K t -1) or the intermediate information of the highest layer among all layers of intermediate information transmitted in the first transmission mode is transmitted to Layer 0 at time t+1, or is transmitted to the lowest layer of all layers of intermediate information transmitted in the first transmission mode at time t+1.

[0107] It should be noted that the above-mentioned transfer of intermediate information to a certain layer can be understood as transferring the intermediate information to the AI ​​unit corresponding to the layer for processing. In addition, the above-mentioned various transfer methods can be combined in any way, and the following examples are illustrated with reference to the accompanying figures:

[0108] Example 1: Some layers transfer intermediate information in the first transfer mode, that is, the intermediate information of Layer x at time t is transferred to Layer (x+1) at time t; some layers transfer intermediate information in the third transfer mode, that is, the intermediate information of Layer x at time t is transferred to Layer (x+1) at time t, and the intermediate information of Layer (K t -1) is passed to Layer 0 at time t+1. For example, as shown in Figure 4a, Layer 0 to Layer 2 at each time point pass the intermediate information according to the first transfer mode, and Layer 3 at each time point except the last time point passes the intermediate information according to the third transfer mode.

[0109] Example 2: Some layers transmit intermediate information using the second transmission method, that is, the intermediate information of Layer x at time t is transmitted to Layer x at time t+1. For example, as shown in Figure 4b, the layers at all moments except the last moment transmit intermediate information using the second transmission method.

[0110] Example 3: Some low layers transmit intermediate information in the first transmission mode, some low layers transmit intermediate information in the third mode, and some high layers transmit intermediate information in the second transmission mode. For example, as shown in Figure 4c and Figure 4d, in Figure 4c, the low layers include Layer0 and Layer1, and the high layers include Layer2 and Layer3; in Figure 4d, the low layers include Layer0, Layer1 and Layer2, and the high layers include Layer3.

[0111] Example 4: Some lower layers transmit intermediate information according to the second transmission method, some higher layers transmit intermediate information according to the first transmission method, and some higher layers transmit intermediate information according to the third transmission method. For example, as shown in Figure 4e, in Figure 4e, the above-mentioned lower layers include Layer0 and Layer1, and the above-mentioned higher layers include Layer2 and Layer3.

[0112] Example 5: Some layers transmit intermediate information according to both the first transmission mode and the second transmission mode, some layers transmit intermediate information according to the second transmission mode, and some layers transmit intermediate information according to the first transmission mode, for example, as shown in FIG4f .

[0113] Example 6: Some layers transmit intermediate information according to the first transmission mode, and some layers transmit intermediate information according to both the first transmission mode and the second transmission mode. For example, as shown in Figure 4g, in Figure 4g, Layer 0 and Layer 1 transmit intermediate information according to both the first transmission mode and the second transmission mode, and Layer 2 transmits intermediate information according to the first transmission mode.

[0114] Example 7: Some layers transmit intermediate information according to the first transmission mode, and some layers transmit intermediate information according to both the first transmission mode and the second transmission mode, for example, as shown in FIG4h , where the values ​​of the three ranks shown in FIG4h are different.

[0115] Example 8: Some layers transmit intermediate information according to the first transmission mode, and some layers transmit intermediate information according to both the first transmission mode and the second transmission mode. For example, as shown in FIG4i , among the three ranks shown in FIG4i , some ranks have the same value, while some ranks have different values.

[0116] Example 9: Some layers transmit intermediate information according to the first transmission method, and some layers transmit intermediate information according to the third transmission method, for example, as shown in Figure 4j, where some ranks of the three ranks shown in Figure 4j have the same values ​​and some ranks have different values.

[0117] Example 10: Some layers transmit intermediate information according to the first transmission method, some layers transmit intermediate information according to the third transmission method, and some layers transmit intermediate information according to the second transmission method, for example, as shown in Figure 4k, where some of the three ranks shown in Figure 4k have the same values ​​and some have different values.

[0118] Example 11: Some layers transmit intermediate information according to the first transmission method, some layers transmit intermediate information according to the third transmission method, and some layers transmit intermediate information according to the second transmission method, for example, as shown in Figure 41, wherein the values ​​of some ranks among the three ranks shown in Figure 41 are the same, and the values ​​of some ranks are different.

[0119] It should be noted that the arrows in FIG. 4 a to FIG. 4 l above indicate the transmission of intermediate information.

[0120] Optionally, in a case where the second target layer includes a fourth layer, the associated layer of the fourth layer includes at least one of the second layer and the third layer;

[0121] or,

[0122] In a case where the second target layer includes a fifth layer, an associated layer of the fifth layer includes at least one of the first layer and the second layer;

[0123] The fourth layer is the lowest layer or the layer with the smallest index among all layers of the second rank, or the fourth layer is the lowest layer or the layer with the smallest index among all layers of the second rank that transmit intermediate information to a higher layer;

[0124] The fifth layer is a layer different from the fourth layer among all layers of the second rank, or the fifth layer is a layer different from the fourth layer among all layers of the second rank that transmit intermediate information to a higher layer.

[0125] Illustratively, as shown in FIG4 a , the fourth layer is Layer 0, and the fifth layer includes Layer 1 to Layer 3; or, as shown in FIG4 e , the fourth layer is Layer 2, and the fifth layer is Layer 3.

[0126] Optionally, the AI ​​unit corresponding to the third target layer of the fifth rank among the N ranks performs processing based on the third information corresponding to the third target layer;

[0127] The fifth rank is the rank corresponding to the fourth time unit among the N ranks, the third information corresponding to the third target layer includes channel information of the third target layer and intermediate information of the associated layer of the third target layer, and the intermediate information of the associated layer of the third target layer is output information of the AI ​​unit corresponding to the associated layer of the third target layer;

[0128] The associated layers of the third target layer include at least one of the following: the sixth layer, the seventh layer, and the eighth layer;

[0129] The sixth layer is a layer of the fifth rank, and the sixth layer is one layer higher than the third target layer, or the index of the sixth layer is the index of the third target layer plus 1;

[0130] The seventh layer is a layer of a sixth rank, the sixth rank being a rank among the N ranks corresponding to a fifth time unit, the fifth time unit being a time unit before the fourth time unit, and an index of the seventh layer is the same as an index of the third target layer;

[0131] The seventh layer is the lowest layer or the layer with the smallest index among all layers of the seventh rank, or the lowest layer or the layer with the smallest index among all layers in which the corresponding AI unit in the layer of the seventh rank uses the intermediate information of a higher layer for processing. The seventh rank is the rank among the N ranks corresponding to the sixth time unit, and the sixth time unit is the time unit before the fourth time unit.

[0132] In this embodiment, the fifth rank may be any rank or a specific rank of the N ranks, and the third target layer may be any layer or a specific layer of the fifth rank.

[0133] The fourth time unit may be any time unit, and the time unit may include but is not limited to a time slot, a sub-time slot, a frame or a sub-frame.

[0134] The fifth time unit is a time unit before the fourth time unit. Exemplarily, the fifth time unit may be the K3 time units before the fourth time unit, where K3 is a positive integer, such as 1 or 2. In some optional embodiments, the fifth time unit is a time unit in which the AI ​​unit was most recently used to process channel information before the fourth time unit, that is, the time unit in which the AI ​​unit was last used, or a time unit in which the same channel information is measured that is closest to the time unit in which the current channel information is measured.

[0135] The sixth time unit and the fifth time unit may be the same time unit or different time units. Exemplarily, the fifth time unit is the K4 time units before the fourth time unit, where K4 is a positive integer, such as 1 or 2. In some optional embodiments, the sixth time unit may be the time unit in which the AI ​​unit was most recently used to process channel information before the fourth time unit, that is, the time unit in which the AI ​​unit was last used, or the time unit in which the same channel information was measured that was closest to the time unit in which the current channel information was measured.

[0136] It can be understood that, when the fifth time unit and the sixth time unit are the same time unit, the sixth rank and the sixth rank are the same rank.

[0137] It should be noted that, when the fifth rank includes multiple ranks, the types of associated layers of the third target layers of different ranks in the multiple ranks may be the same, or may be different, wherein the types of associated layers include the sixth layer, the seventh layer, and the eighth layer. For example, the associated layer of the third target layer of some ranks is the sixth layer, the associated layer of the third target layer of some ranks is the eighth layer, and the associated layers of the third target layer of some ranks include the sixth layer and the seventh layer, etc.

[0138] It should also be noted that, in the case where the third target layer includes multiple layers, the AI ​​unit corresponding to each of the multiple layers performs processing based on the third information corresponding to the layer. In addition, the types of associated layers of different layers in the multiple layers may be different, for example, the associated layer of some layers is the sixth layer, the associated layer of some layers is the seventh layer, the associated layer of some layers includes the sixth layer and the seventh layer, etc.; or the types of associated layers of different layers in the multiple layers may be the same, for example, the associated layer of each of the multiple layers is the sixth layer, or is the seventh layer, or includes the sixth layer and the seventh layer, etc. For example, the above-mentioned third target layer includes layer 1, layer 2 and layer 3 of the fourth time unit, wherein the associated layer of layer 1 of the fourth time unit may include the sixth layer, that is, the associated layer of layer 1 of the fourth time unit is layer 2 of the fourth time unit; the associated layer of layer 2 of the fourth time unit may include the sixth layer and the seventh layer, that is, the associated layer of layer 2 of the fourth time unit includes layer 3 of the fourth time unit and layer 2 of the fifth time unit; the associated layer of layer 3 of the fourth time unit may include the seventh layer, that is, the associated layer of layer 3 of the fourth time unit is layer 3 of the fifth time unit.

[0139] For example, the intermediate information of the sixth layer can be understood as the intermediate information transmitted in accordance with the fourth transmission mode, wherein the fourth transmission mode is: transmitting the intermediate information of Layer x at time t to Layer (x-1) at time t, x <K t -1,K t Indicates the value of the rank corresponding to time t.

[0140] The intermediate information of the seventh layer can be understood as the intermediate information transmitted in the second transmission mode, wherein the second transmission mode is: the intermediate information of Layer x at time t is transmitted to Layer x at time t+1 or Layer x at the time closest to time t. <K t -1 and x <K t+1 -1, the above K t+1 Indicates the value of the rank corresponding to time t+1.

[0141] The intermediate information of the eighth layer can be understood as the intermediate information transmitted in accordance with the fifth transmission mode, wherein the fifth transmission mode is: the intermediate information of Layer 0 at time t or the lowest layer of all layers of the intermediate information transmitted in accordance with the fourth transmission mode is transmitted to Layer K at time t+1. t+1 -1, or the highest layer of all layers that transmit intermediate information according to the fourth transmission mode at time t+1, where K t+1 Indicates the value of the rank at time t+1.

[0142] It should be noted that the above-mentioned transfer of intermediate information to a certain layer can be understood as transferring the intermediate information to the AI ​​unit corresponding to the layer for processing. In addition, the above-mentioned various transfer methods can be combined in any way, and the following examples are illustrated with reference to the accompanying figures:

[0143] Example 1: Some layers pass intermediate information using the fourth transfer method, that is, the intermediate information of Layer x at time t is passed to Layer x-1 at time t. Some layers pass intermediate information using the fifth transfer method, that is, the intermediate information of Layer 0 at time t is passed to Layer K at time t+1. t+1 -1, for example, as shown in FIG5a, Layer 1 to Layer 3 at each moment transmit intermediate information according to the fourth transmission mode, and Layer 0 at each moment except the last moment transmits intermediate information according to the fifth transmission mode.

[0144] Example 2: Some low layers transmit intermediate information according to the fourth transmission method, some low layers transmit intermediate information according to the fifth transmission method, and some high layers transmit intermediate information according to the second transmission method, for example, as shown in Figure 5b and Figure 5c, wherein, in Figure 5b, the above-mentioned low layers include Layer0 and Layer1, and the above-mentioned high layers include Layer2 and Layer3; in Figure 5c, the above-mentioned low layers include Layer0, Layer1 and Layer2, and the above-mentioned high layers include Layer3.

[0145] Example 3: Some lower layers transmit intermediate information according to the second transmission method, some higher layers transmit intermediate information according to the fourth transmission method, and some higher layers transmit intermediate information according to the fifth transmission method. For example, as shown in Figure 5d, in Figure 5d, the above-mentioned lower layers include Layer0 and Layer1, and the above-mentioned higher layers include Layer2 and Layer3.

[0146] Example 4: Some layers transmit intermediate information according to both the second transmission mode and the fourth transmission mode. Some layers transmit intermediate information according to the second transmission mode, and some layers transmit intermediate information according to the fourth transmission mode, for example, as shown in FIG5e.

[0147] Example 5: Some layers transmit intermediate information according to the fourth transmission mode, and some layers transmit intermediate information according to both the fourth transmission mode and the second transmission mode. For example, as shown in Figure 5f, in Figure 5f, Layer 0 and Layer 1 transmit intermediate information according to both the fourth transmission mode and the second transmission mode, and Layer 2 and Layer 3 transmit intermediate information according to the fourth transmission mode.

[0148] Example 6: Some layers transmit intermediate information according to the fourth transmission mode, some layers transmit intermediate information according to both the fourth transmission mode and the second transmission mode, and some layers transmit intermediate information according to the second transmission mode, for example, as shown in FIG5g, wherein the values ​​of the three ranks shown in FIG5g are all different.

[0149] Example 7: Some layers transmit intermediate information according to the fourth transmission mode and the second transmission mode at the same time, some layers transmit intermediate information according to the second transmission mode, and some layers transmit intermediate information according to the fourth transmission mode, for example, as shown in Figure 5h, wherein the values ​​of some ranks among the three ranks shown in Figure 5h are the same, and the values ​​of some ranks are different.

[0150] Example 8: Some layers transmit intermediate information according to the fourth transmission method, and some layers transmit intermediate information according to the fifth transmission method, for example, as shown in Figure 5i, where some ranks of the three ranks shown in Figure 5i have the same values ​​and some ranks have different values.

[0151] Example nine: Some layers transmit intermediate information according to the fourth transmission method, some layers transmit intermediate information according to the fifth transmission method, and some layers transmit intermediate information according to the second transmission method, for example, as shown in Figure 5j, wherein the values ​​of some ranks of the three ranks shown in Figure 5j are the same, and the values ​​of some ranks are different.

[0152] It should be noted that the arrows in FIG. 5 a to FIG. 5 j above indicate the transmission paths of the intermediate information.

[0153] Optionally, in a case where the third target layer includes a ninth layer, the associated layer of the ninth layer includes at least one of the seventh layer and the eighth layer;

[0154] or,

[0155] In a case where the third target layer includes the tenth layer, the associated layer of the tenth layer includes at least one of the sixth layer and the seventh layer;

[0156] The ninth layer is the topmost layer or the layer with the largest index among all layers of the fifth rank, or the ninth layer is the topmost layer or the layer with the largest index among all layers of the fifth rank that transmit intermediate information to a lower layer;

[0157] The tenth layer is a layer different from the ninth layer among all layers of the fifth rank, or the tenth layer is a layer different from the ninth layer among all layers of the fifth rank that transmit intermediate information to a higher layer.

[0158] Exemplarily, as shown in FIG5 a , the ninth layer is Layer 0, and the tenth layer includes Layer 1 to Layer 3; or, as shown in FIG5 d , the ninth layer is Layer 2, and the tenth layer is Layer 3.

[0159] In summary, the embodiments of the present application provide a method for combining single-layer AI units into high-rank (i.e., rank greater than 1) AI units, as well as a method for the interaction of intermediate information of each layer between AI units corresponding to different layers, thereby optimizing the AI ​​processing method of high-rank space-time-frequency domain compression, so that both the terminal and the network side can better align the AI ​​reasoning method and the interaction process of intermediate information, thereby ensuring the performance of space-time-frequency domain compression.

[0160] It should be noted that the information processing method provided in the embodiments of the present application can be executed by an information processing device, or by a control module in the information processing device for executing the information processing method. In the embodiments of the present application, the information processing device provided in the embodiments of the present application is described by taking the information processing device executing the information processing method as an example.

[0161] Please refer to FIG6 , which is a structural diagram of an information processing device provided in an embodiment of the present application. As shown in FIG6 , the information processing device 600 includes:

[0162] A processing module 601 is configured to use artificial intelligence (AI) units corresponding to N ranks to process channel information of each layer of the N ranks respectively;

[0163] The AI ​​unit corresponding to each of the ranks includes the AI ​​units corresponding to all layers of each of the ranks, the N ranks include at least one rank greater than 1, and N is a positive integer.

[0164] Optionally, different ranks among the N ranks correspond to different AI units, and the same rank among the N ranks corresponds to the same AI unit.

[0165] Optionally, different layers of each rank correspond to different AI units.

[0166] Optionally, the same layer in different ranks among the N ranks corresponds to the same AI unit, and different layers of the same rank among the N ranks correspond to different AI units.

[0167] Optionally, the same layer in different ranks among the N ranks corresponds to the same AI unit, and different layers of the same rank among the N ranks correspond to the same AI unit.

[0168] Optionally, the AI ​​unit corresponding to the first target layer of the first rank of the N ranks performs processing based on the first information corresponding to the first target layer;

[0169] Among them, the first information corresponding to the first target layer includes channel information of the first target layer and intermediate information of the associated layer of the first target layer, the associated layer of the first target layer includes at least one layer of all layers of the N ranks except the first target layer, and the intermediate information of the associated layer of the first target layer is the output information of the AI ​​unit corresponding to the associated layer of the first target layer.

[0170] Optionally, the intermediate information of the associated layer of the first target layer is the input information of the input layer of the AI ​​unit corresponding to the first target layer, or the intermediate information of the associated layer of the first target layer is the input information of at least one hidden layer of the AI ​​unit corresponding to the first target layer.

[0171] Optionally, the AI ​​unit corresponding to the second target layer of the second rank among the N ranks performs processing based on the second information corresponding to the second target layer;

[0172] The second rank is the rank corresponding to the first time unit among the N ranks, the second information corresponding to the second target layer includes channel information of the second target layer and intermediate information of an associated layer of the second target layer, and the intermediate information of the associated layer of the second target layer is output information of the AI ​​unit corresponding to the associated layer of the second target layer;

[0173] The associated layers of the second target layer include at least one of the following: a first layer, a second layer, and a third layer;

[0174] The first layer is a layer of the second rank, and the first layer is one layer lower than the second target layer, or the index of the first layer is the index of the second target layer minus 1;

[0175] The second layer is a layer of a third rank, the third rank being a rank among the N ranks corresponding to a second time unit, the second time unit being a time unit before the first time unit, and an index of the second layer is the same as an index of the second target layer;

[0176] The third layer is the topmost layer or the layer with the largest index among all layers of the fourth rank, or the third layer is the topmost layer or the layer with the largest index among all layers whose corresponding AI unit in the layer of the fourth rank uses the intermediate information of the lower layer for processing, and the fourth rank is the rank corresponding to the third time unit among the N ranks, and the third time unit is the time unit before the first time unit.

[0177] Optionally, in a case where the second target layer includes a fourth layer, the associated layer of the fourth layer includes at least one of the second layer and the third layer;

[0178] or,

[0179] In a case where the second target layer includes a fifth layer, an associated layer of the fifth layer includes at least one of the first layer and the second layer;

[0180] The fourth layer is the lowest layer or the layer with the smallest index among all layers of the second rank, or the fourth layer is the lowest layer or the layer with the smallest index among all layers of the second rank that transmit intermediate information to a higher layer;

[0181] The fifth layer is a layer different from the fourth layer among all layers of the second rank, or the fifth layer is a layer different from the fourth layer among all layers of the second rank that transmit intermediate information to a higher layer.

[0182] Optionally, the AI ​​unit corresponding to the third target layer of the fifth rank among the N ranks performs processing based on the third information corresponding to the third target layer;

[0183] The fifth rank is the rank corresponding to the fourth time unit among the N ranks, the third information corresponding to the third target layer includes channel information of the third target layer and intermediate information of the associated layer of the third target layer, and the intermediate information of the associated layer of the third target layer is output information of the AI ​​unit corresponding to the associated layer of the third target layer;

[0184] The associated layers of the third target layer include at least one of the following: the sixth layer, the seventh layer, and the eighth layer;

[0185] The sixth layer is a layer of the fifth rank, and the sixth layer is one layer higher than the third target layer, or the index of the sixth layer is the index of the third target layer plus 1;

[0186] The seventh layer is a layer of a sixth rank, the sixth rank being a rank among the N ranks corresponding to a fifth time unit, the fifth time unit being a time unit before the fourth time unit, and an index of the seventh layer is the same as an index of the third target layer;

[0187] The seventh layer is the lowest layer or the layer with the smallest index among all layers of the seventh rank, or the lowest layer or the layer with the smallest index among all layers in which the corresponding AI unit in the layer of the seventh rank uses the intermediate information of a higher layer for processing. The seventh rank is the rank among the N ranks corresponding to the sixth time unit, and the sixth time unit is the time unit before the fourth time unit.

[0188] Optionally, in a case where the third target layer includes a ninth layer, the associated layer of the ninth layer includes at least one of the seventh layer and the eighth layer;

[0189] or,

[0190] In a case where the third target layer includes the tenth layer, the associated layer of the tenth layer includes at least one of the sixth layer and the seventh layer;

[0191] The ninth layer is the topmost layer or the layer with the largest index among all layers of the fifth rank, or the ninth layer is the topmost layer or the layer with the largest index among all layers of the fifth rank that transmit intermediate information to a lower layer;

[0192] The tenth layer is a layer different from the ninth layer among all layers of the fifth rank, or the tenth layer is a layer different from the ninth layer among all layers of the fifth rank that transmit intermediate information to a higher layer.

[0193] The information processing device in the embodiments of the present application 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 chip. The electronic device can be a terminal or a network-side device, or can be a device other than a terminal and a network-side device. For example, the terminal can include but is not limited to the types of terminals 11 listed above, the network-side device can include but is not limited to the types of network-side devices 12 listed above, and other devices can be servers, network attached storage (NAS), etc., which are not specifically limited in the embodiments of the present application.

[0194] The information processing device provided in the embodiment of the present application can implement each process implemented in the method embodiment of Figure 3 and achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0195] Optionally, as shown in Figure 7, an embodiment of the present application also provides a communication device 700, including a processor 701 and a memory 702, and the memory 702 stores a program or instruction that can be run on the processor 701. When the program or instruction is executed by the processor 701, the various steps of the above-mentioned information processing method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0196] An embodiment of the present application also provides a terminal, including a processor and a communication interface, wherein the processor is configured to use artificial intelligence AI units corresponding to N ranks to process the channel information of each layer of the N ranks respectively; wherein the AI ​​units corresponding to each of the ranks respectively include AI units corresponding to all layers of each of the ranks, the N ranks include at least one rank greater than 1, and N is a positive integer. Each implementation process and implementation method of the above-mentioned method embodiment can be applied to the terminal embodiment and can achieve the same technical effect. Specifically, Figure 8 is a schematic diagram of the hardware structure of a terminal that implements an embodiment of the present application.

[0197] The terminal 800 includes but is not limited to: a radio frequency unit 801, a network module 802, an audio output unit 803, an input unit 804, a sensor 805, a display unit 806, a user input unit 807, an interface unit 808, a memory 809 and at least some of the components of the processor 810.

[0198] Those skilled in the art will appreciate that the terminal 800 may also include a power supply (such as a battery) to power various components. The power supply may be logically connected to the processor 810 through a power management system, thereby implementing functions such as charging, discharging, and power consumption management through the power management system. The terminal structure shown in FIG8 does not constitute a limitation of the terminal. The terminal may include more or fewer components than shown, or combine certain components, or arrange the components differently, which will not be described in detail here.

[0199] It should be understood that in an embodiment of the present application, the input unit 804 may include a graphics processing unit (GPU) 8041 and a microphone 8042, and the graphics processor 8041 processes the image data of a static picture or video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The display unit 806 may include a display panel 8061, and the display panel 8061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 807 includes a touch panel 8071 and at least one of other input devices 8072. The touch panel 8071 is also called a touch screen. The touch panel 8071 may include two parts: a touch detection device and a touch controller. Other input devices 8072 may include, but are not limited to, a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.

[0200] In the embodiment of the present application, after receiving downlink data from a network-side device, the radio frequency unit 801 may transmit the data to the processor 810 for processing. Furthermore, the radio frequency unit 801 may send uplink data to the network-side device. Typically, the radio frequency unit 801 includes, but is not limited to, an antenna, an amplifier, a transceiver, a coupler, a low-noise amplifier, a duplexer, and the like.

[0201] The memory 809 can be used to store software programs or instructions and various data. The memory 809 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, applications or instructions required for at least one function (such as a sound playback function, an image playback function, etc.). In addition, the memory 809 may include a volatile memory or a non-volatile memory. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may 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 RAM bus random access memory (DRRAM). The memory 809 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.

[0202] Processor 810 may include one or more processing units. Optionally, processor 810 integrates an application processor and a modem processor. The application processor primarily handles operations related to the operating system, user interface, and application programs, while the modem processor primarily processes wireless communication signals, such as a baseband processor. It is understood that the modem processor may not be integrated into processor 810.

[0203] Among them, the processor 810 is used to use artificial intelligence AI units corresponding to N ranks to process the channel information of each layer of the N ranks respectively; wherein the AI ​​unit corresponding to each of the ranks respectively includes the AI ​​units corresponding to all layers of each of the ranks, the N ranks include at least one rank greater than 1, and N is a positive integer.

[0204] It can be understood that the implementation process of each implementation method mentioned in this embodiment can refer to the relevant description of the aforementioned method embodiment and achieve the same or corresponding technical effects. To avoid repetition, it will not be repeated here.

[0205] An embodiment of the present application also provides a network-side device, including a processor and a communication interface, wherein the processor is configured to use artificial intelligence (AI) units corresponding to N ranks to process channel information of each layer of the N ranks; wherein the AI ​​units corresponding to each rank include AI units corresponding to all layers of each rank, the N ranks include at least one rank greater than 1, and N is a positive integer. The various implementation processes and implementation methods of the above-mentioned method embodiment are applicable to the network-side device embodiment and can achieve the same technical effects.

[0206] Specifically, embodiments of the present application also provide a network-side device. As shown in Figure 9, the network-side device 900 includes an antenna 901, a radio frequency device 902, a baseband device 903, a processor 904, and a memory 905. Antenna 901 is connected to radio frequency device 902. In the uplink direction, radio frequency device 902 receives information via antenna 901 and sends the received information to baseband device 903 for processing. In the downlink direction, baseband device 903 processes the information to be transmitted and sends it to radio frequency device 902. Radio frequency device 902 processes the received information and then sends it through antenna 901.

[0207] The method executed by the network-side device in the above embodiment may be implemented in the baseband device 903 , which includes a baseband processor.

[0208] The baseband device 903 may include, for example, at least one baseband board, on which multiple chips are arranged, as shown in Figure 9, one of which is, for example, a baseband processor, which is connected to the memory 905 through a bus interface to call the program in the memory 905 and execute the network device operations shown in the above method embodiment.

[0209] The network side device may further include a network interface 906, which is, for example, a Common Public Radio Interface (CPRI).

[0210] Specifically, the network side device 900 of the embodiment of the present application also includes: instructions or programs stored in the memory 905 and can be run on the processor 904. The processor 904 calls the instructions or programs in the memory 905 to execute the method of execution of each module shown in Figure 6 and achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0211] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned information processing method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0212] The processor is the processor in the terminal described in the above embodiment. The readable storage medium includes a computer-readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk. In some examples, the readable storage medium may be a non-transitory readable storage medium.

[0213] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned information processing method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0214] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0215] An embodiment of the present application further provides a computer program / program product, which is stored in a storage medium. The computer program / program product is executed by at least one processor to implement the various processes of the above-mentioned information processing method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0216] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the opposite order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may also be added, omitted or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0217] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of a computer software product plus a necessary general-purpose hardware platform, or of course, by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disk, optical disk, etc.) and includes a number of instructions for enabling a terminal or network-side device to execute the methods described in each embodiment of the present application.

[0218] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms of implementation methods without departing from the purpose of this application and the scope of protection of the claims. These implementation methods are all within the protection of this application.

Claims

1. An information processing method, comprising: A communication device processes channel information of each layer of the N ranks by using artificial intelligence (AI) units corresponding to the N ranks respectively; Wherein, each AI unit corresponding to a rank respectively includes AI units corresponding to all layers of each rank, the N ranks include at least one rank greater than 1, and N is a positive integer.

2. The method according to claim 1, wherein, The AI units corresponding to different ranks among the N ranks are different, and the AI units corresponding to the same rank among the N ranks are the same.

3. The method according to claim 1 or 2, wherein Different layers of each rank correspond to different AI units.

4. The method according to claim 1, wherein, The same layer in different ranks among the N ranks corresponds to the same AI unit, and different layers of the same rank among the N ranks correspond to different AI units.

5. The method according to claim 1, wherein, The same layer in different ranks among the N ranks corresponds to the same AI unit, and different layers of the same rank among the N ranks correspond to the same AI unit.

6. The method according to any one of claims 1 to 5, wherein The AI unit corresponding to the first target layer of the first rank among the N ranks processes based on the first information corresponding to the first target layer; Wherein, the first information corresponding to the first target layer includes the channel information of the first target layer and the intermediate information of the associated layer of the first target layer, the associated layer of the first target layer includes at least one layer other than the first target layer among all layers of the N ranks, and the intermediate information of the associated layer of the first target layer is the output information of the AI unit corresponding to the associated layer of the first target layer.

7. The method according to claim 6, wherein The intermediate information of the associated layer of the first target layer is the input information of the input layer of the AI unit corresponding to the first target layer, or the intermediate information of the associated layer of the first target layer is the input information of at least one hidden layer of the AI unit corresponding to the first target layer.

8. The method according to any one of claims 1 to 7, wherein The AI unit corresponding to the second target layer of the second rank among the N ranks processes based on the second information corresponding to the second target layer; Wherein, the second rank is the rank corresponding to the first time unit among the N ranks, the second information corresponding to the second target layer includes the channel information of the second target layer and the intermediate information of the associated layer of the second target layer, and the intermediate information of the associated layer of the second target layer is the output information of the AI unit corresponding to the associated layer of the second target layer; The associated layer of the second target layer includes at least one of the following: the first layer, the second layer, and the third layer; The first layer is a layer of the second rank, and the first layer is the layer lower than the second target layer, or the index of the first layer is the index of the second target layer minus 1; The second layer is a layer of the third rank, the third rank is the rank corresponding to the second time unit among the N ranks, the second time unit is the time unit before the first time unit, and the index of the second layer is the same as the index of the second target layer; The third layer is the top layer or the layer with the largest index among all layers of the fourth rank, or the third layer is the top layer or the layer with the largest index among all layers of the fourth rank where the corresponding AI unit of the fourth rank processes by using the intermediate information of the lower layer, the fourth rank is the rank corresponding to the third time unit among the N ranks, and the third time unit is the time unit before the first time unit.

9. The method according to claim 8, wherein When the second target layer includes the fourth layer, the associated layer of the fourth layer includes at least one of the second layer and the third layer; Or, When the second target layer includes the fifth layer, the associated layer of the fifth layer includes at least one of the first layer and the second layer; Wherein, the fourth layer is the bottom layer or the layer with the smallest index among all the layers of the second rank, or the fourth layer is the bottom layer or the layer with the smallest index among all the layers of the second rank that transmit intermediate information to the layer one level higher than it; The fifth layer is a layer different from the fourth layer among all the layers of the second rank, or the fifth layer is a layer different from the fourth layer among all the layers of the second rank that transmit intermediate information to the layer one level higher than it.

10. The method according to any one of claims 1 to 9, wherein, The AI unit corresponding to the third target layer of the fifth rank among the N ranks processes based on the third information corresponding to the third target layer; Wherein, the fifth rank is the rank corresponding to the fourth time unit among the N ranks, the third information corresponding to the third target layer includes the channel information of the third target layer and the intermediate information of the associated layer of the third target layer, and the intermediate information of the associated layer of the third target layer is the output information of the AI unit corresponding to the associated layer of the third target layer; The associated layer of the third target layer includes at least one of the following: the sixth layer, the seventh layer, and the eighth layer; The sixth layer is a layer of the fifth rank, and the sixth layer is the layer one level higher than the third target layer, or the index of the sixth layer is the index of the third target layer plus 1; The seventh layer is a layer of the sixth rank, the sixth rank is the rank corresponding to the fifth time unit among the N ranks, the fifth time unit is the time unit before the fourth time unit, and the index of the seventh layer is the same as the index of the third target layer; The seventh layer is the bottom layer or the layer with the smallest index among all the layers of the seventh rank, or the bottom layer or the layer with the smallest index among all the layers of the seventh rank where the corresponding AI unit processes using the intermediate information of the layer one level higher, the seventh rank is the rank corresponding to the sixth time unit among the N ranks, and the sixth time unit is the time unit before the fourth time unit.

11. The method according to claim 10, wherein, When the third target layer includes the ninth layer, the associated layer of the ninth layer includes at least one of the seventh layer and the eighth layer; Or, When the third target layer includes the tenth layer, the associated layer of the tenth layer includes at least one of the sixth layer and the seventh layer; Wherein, the ninth layer is the top layer or the layer with the largest index among all the layers of the fifth rank, or the ninth layer is the top layer or the layer with the largest index among all the layers of the fifth rank that transmit intermediate information to the layer one level lower than it; The tenth layer is a layer different from the ninth layer among all the layers of the fifth rank, or the tenth layer is a layer different from the ninth layer among all the layers of the fifth rank that transmit intermediate information to the layer one level higher than it.

12. An information processing device, comprising: A processing module, which is configured to process the channel information of each layer of the N ranks respectively by using artificial intelligence (AI) units corresponding to the N ranks; Wherein, each AI unit corresponding to a rank respectively includes AI units corresponding to all layers of each rank, the N ranks include at least one rank greater than 1, and N is a positive integer.

13. The apparatus according to claim 12, wherein, The AI units corresponding to different ranks among the N ranks are different, and the AI units corresponding to the same rank among the N ranks are the same.

14. The apparatus according to claim 12 or 13, wherein, Different layers of each rank correspond to different AI units.

15. The apparatus according to claim 12, wherein, The same layer in different ranks among the N ranks corresponds to the same AI unit, and different layers of the same rank among the N ranks correspond to different AI units.

16. The apparatus according to claim 12, wherein, The same layer in different ranks among the N ranks corresponds to the same AI unit, and different layers of the same rank among the N ranks correspond to the same AI unit.

17. The device according to any one of claims 12 to 16, wherein, The AI unit corresponding to the first target layer of the first rank among the N ranks processes based on the first information corresponding to the first target layer; Wherein, the first information corresponding to the first target layer includes the channel information of the first target layer and the intermediate information of the associated layer of the first target layer, the associated layer of the first target layer includes at least one layer other than the first target layer among all layers of the N ranks, and the intermediate information of the associated layer of the first target layer is the output information of the AI unit corresponding to the associated layer of the first target layer.

18. The apparatus according to claim 17, wherein The intermediate information of the associated layer of the first target layer is the input information of the input layer of the AI unit corresponding to the first target layer, or the intermediate information of the associated layer of the first target layer is the input information of at least one hidden layer of the AI unit corresponding to the first target layer.

19. The device according to any one of claims 12 to 18, wherein, The AI unit corresponding to the second target layer of the second rank among the N ranks processes based on the second information corresponding to the second target layer; Wherein, the second rank is the rank corresponding to the first time unit among the N ranks, the second information corresponding to the second target layer includes the channel information of the second target layer and the intermediate information of the associated layer of the second target layer, and the intermediate information of the associated layer of the second target layer is the output information of the AI unit corresponding to the associated layer of the second target layer; The associated layer of the second target layer includes at least one of the following: the first layer, the second layer, and the third layer; The first layer is a layer of the second rank, and the first layer is the layer lower than the second target layer, or the index of the first layer is the index of the second target layer minus 1; The second layer is a layer of the third rank, the third rank is the rank corresponding to the second time unit among the N ranks, the second time unit is the time unit before the first time unit, and the index of the second layer is the same as the index of the second target layer; The third layer is the top layer or the layer with the largest index among all layers of the fourth rank, or the third layer is the top layer or the layer with the largest index among all layers of the fourth rank where the corresponding AI unit of the fourth rank processes by using the intermediate information of the lower layer, the fourth rank is the rank corresponding to the third time unit among the N ranks, and the third time unit is the time unit before the first time unit.

20. The apparatus according to claim 19, wherein When the second target layer includes the fourth layer, the associated layer of the fourth layer includes at least one of the second layer and the third layer; Or, When the second target layer includes the fifth layer, the associated layer of the fifth layer includes at least one of the first layer and the second layer; Wherein, the fourth layer is the bottom layer or the layer with the smallest index among all the layers of the second rank, or, the fourth layer is the bottom layer or the layer with the smallest index among all the layers of the second rank that transmit intermediate information to the layer one level higher than it; The fifth layer is a layer different from the fourth layer among all the layers of the second rank, or, the fifth layer is a layer different from the fourth layer among all the layers of the second rank that transmit intermediate information to the layer one level higher than it.

21. The device according to any one of claims 12 to 20, wherein, The AI unit corresponding to the third target layer of the fifth rank among the N ranks processes based on the third information corresponding to the third target layer; Wherein, the fifth rank is the rank corresponding to the fourth time unit among the N ranks, the third information corresponding to the third target layer includes the channel information of the third target layer and the intermediate information of the associated layer of the third target layer, and the intermediate information of the associated layer of the third target layer is the output information of the AI unit corresponding to the associated layer of the third target layer; The associated layer of the third target layer includes at least one of the following: the sixth layer, the seventh layer, the eighth layer; The sixth layer is a layer of the fifth rank, and the sixth layer is the layer one level higher than the third target layer, or, the index of the sixth layer is the index of the third target layer plus 1; The seventh layer is a layer of the sixth rank, the sixth rank is the rank corresponding to the fifth time unit among the N ranks, the fifth time unit is the time unit before the fourth time unit, and the index of the seventh layer is the same as the index of the third target layer; The seventh layer is the bottom layer or the layer with the smallest index among all the layers of the seventh rank, or, the bottom layer or the layer with the smallest index among all the layers of the seventh rank where the corresponding AI unit processes using the intermediate information of the layer one level higher, the seventh rank is the rank corresponding to the sixth time unit among the N ranks, and the sixth time unit is the time unit before the fourth time unit.

22. The apparatus according to claim 21, wherein, When the third target layer includes the ninth layer, the associated layer of the ninth layer includes at least one of the seventh layer and the eighth layer; Or, When the third target layer includes the tenth layer, the associated layer of the tenth layer includes at least one of the sixth layer and the seventh layer; Wherein, the ninth layer is the top layer or the layer with the largest index among all the layers of the fifth rank, or, the ninth layer is the top layer or the layer with the largest index among all the layers of the fifth rank that transmit intermediate information to the layer one level lower than it; The tenth layer is a layer different from the ninth layer among all the layers of the fifth rank, or, the tenth layer is a layer different from the ninth layer among all the layers of the fifth rank that transmit intermediate information to the layer one level higher than it.

23. A communication device, comprising a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the programs or instructions are executed by the processor, the steps of the information processing method according to any one of claims 1 to 11 are implemented.

24. A readable storage medium, on which programs or instructions are stored, and when the programs or instructions are executed by a processor, the steps of the information processing method according to any one of claims 1 to 11 are implemented.

Images