WO2022151069A1

WO2022151069A1 - Method and apparatus for processing received information, computer device, and storage medium

Info

Publication number: WO2022151069A1
Application number: PCT/CN2021/071547
Authority: WO
Inventors: 田文强; 肖寒; 刘文东
Original assignee: Oppo广东移动通信有限公司
Priority date: 2021-01-13
Filing date: 2021-01-13
Publication date: 2022-07-21
Also published as: CN116458094A

Abstract

The present application relates to the technical field of wireless communications. Disclosed are a method and apparatus for processing received information, a computer device, and a storage medium. The method comprises: performing wireless signal receiving to obtain received information; acquiring transmission mode information corresponding to the received information; determining a receiver model according to the transmission mode information; and processing the received information by means of the receiver model to obtain first information, the receiver model being a machine learning model obtained by training according to a received information sample and an information processing sample. By means of the solution, for different received information, according to transmission modes of the different received information, a receiving end device can use different receiver models for data processing, thereby improving the accuracy of data processing.

Description

Received information processing method, device, computer equipment and storage medium

technical field

The present application relates to the field of wireless communication technologies, and in particular, to a method, an apparatus, a computer device, and a storage medium for processing received information.

Background technique

In a wireless communication system, in order to realize the normal transmission of data in free space, the transmitting end needs to encode the data to be sent through the transmitter, and send it to the space through the antenna. Correspondingly, the electromagnetic signal received by the antenna of the receiving end is It is sent to the receiver for decoding to realize the normal transmission of data.

In the prior art, the workflow is that the transmitter performs operations such as encoding, modulation, and encryption on the information source at the transmitting end to form the transmission information to be transmitted. The transmitted information is transmitted to the receiving end through the wireless space, and the receiving end decodes, decrypts and demodulates the received received information, and finally restores the source information. That is, there are separate decoding, demodulation, decryption and other modules in the receiver, which correspond to the transmitter's coding, modulation, encryption and other modules, so as to realize the normal processing of data.

In the above solution, each module of the receiver is designed corresponding to each module of the transmitter, and the bit error rate during data processing of the receiver is relatively high.

SUMMARY OF THE INVENTION

Embodiments of the present application provide a received information processing method, apparatus, computer equipment, and storage medium. The technical solution is as follows:

On the one hand, an embodiment of the present application provides a method for processing received information, the method is used for a receiving end device, and the method includes:

Perform wireless signal reception to obtain reception information;

obtaining the transmission mode information corresponding to the received information;

determining a receiver model according to the transmission mode information;

The received information is processed by the receiver model to obtain the first information; the receiver model is a machine learning model obtained after training according to the received information samples and the information processing samples.

In another aspect, an embodiment of the present application provides an apparatus for processing received information, the apparatus is used for a receiving end device, and the apparatus includes:

The wireless signal receiving module is used to perform wireless signal receiving and obtain receiving information;

a transmission mode information acquisition module, configured to acquire the transmission mode information corresponding to the received information;

a model determination module, configured to determine a receiver model according to the transmission mode information;

The first information acquisition module is configured to process the received information through the receiver model to obtain the first information; the receiver model is a machine learning model obtained after training according to the received information samples and the information processing samples.

In yet another aspect, an embodiment of the present application provides a computer device, the computer device is a receiver device, the receiver device includes a processor, a memory, and a transceiver, the memory stores a computer program, and the computer program for being executed by the processor to realize the above-mentioned method for processing received information.

In another aspect, an embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored in the storage medium, and the computer program is loaded and executed by a processor to implement the above method for processing received information.

On the other hand, the embodiments of the present application also provide a computer program product or computer program, where the computer program product or computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the terminal reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the terminal executes the above-mentioned method for processing received information.

The technical solutions provided in the embodiments of the present application can bring the following beneficial effects:

The receiver device uses the machine learning model trained by the training samples as the receiver model, and processes the received received information according to the receiver model to obtain the first information. Through the above scheme, the neural network model is used in the processing method of received information, and according to the different transmission methods, the receiver model used to process the received information is also different, that is, for different received information, according to different received information The transmission method uses different receiver models for data processing, which improves the accuracy of data processing.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the drawings that are used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without creative effort.

FIG. 1 is a schematic diagram of a network architecture of a communication system provided by an embodiment of the present application.

FIG. 2 shows a flow chart of signal transmission in a wireless communication system.

FIG. 3 shows a flowchart of a method for processing received information provided by an embodiment of the present application.

FIG. 4 shows a flowchart of a method for processing received information provided by an embodiment of the present application.

FIG. 5 shows a schematic diagram of a fully connected network involved in the embodiment shown in FIG. 4 .

FIG. 6 shows a schematic diagram of a convolutional neural network involved in the embodiment shown in FIG. 4 .

FIG. 7 shows a receiver neural network model involved in the embodiment shown in FIG. 4 .

FIG. 8 shows a receiver neural network model involved in the embodiment shown in FIG. 4 .

FIG. 9 is a schematic flowchart of a receiving end device implementing a method for processing received information according to an exemplary embodiment of the present application.

FIG. 10 is a schematic flowchart of a receiving end device implementing a method for processing received information according to an exemplary embodiment of the present application.

FIG. 11 shows a block diagram of an apparatus for processing received information provided by an embodiment of the present application.

FIG. 12 shows a schematic structural diagram of a computer device provided by an embodiment of the present application.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

The network architecture and service scenarios described in the embodiments of the present application are for the purpose of illustrating the technical solutions of the embodiments of the present application more clearly, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. The evolution of new business scenarios and the emergence of new business scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

Please refer to FIG. 1 , which shows a schematic diagram of a network architecture of a communication system provided by an embodiment of the present application. The network architecture may include: a terminal 10 and a network-side device 20 .

The number of terminals 10 is usually multiple, and one or more terminals 10 may be distributed in a cell managed by each network-side device (base station) 20 . The terminal 10 may include various handheld devices with wireless communication functions, vehicle-mounted devices, wearable devices, computing devices or other processing devices connected to the wireless modem, as well as various forms of user equipment (User Equipment, UE), mobile stations ( Mobile Station, MS), terminal device, etc. For convenience of description, in the embodiments of the present application, the devices mentioned above are collectively referred to as terminals.

The network-side device 20 is a device deployed in an access network to provide a wireless communication function for the terminal 10 . The network-side device 20 may include various forms of macro-network-side devices, micro-network-side devices, relay stations, access points, and the like. In systems using different wireless access technologies, the names of devices with network-side device functions may be different. For example, in a 5G New Radio (NR) system, the network-side device (ie, base station), It is called gNodeB or gNB. As communication technology evolves, the name "base station" may change. For convenience of description, in the embodiments of the present application, the above-mentioned apparatuses for providing a wireless communication function for the terminal 10 are collectively referred to as network-side devices.

Optionally, what is not shown in FIG. 1 is that the above-mentioned network architecture also includes other network devices, such as: a central control node (Central network control, CNC), a session management function (Session management function, SMF), a user plane function ( User Plane Function, UPF) equipment or AMF (Access and Mobility Management Function, access and mobility management function) equipment and so on.

The "5G NR system" in the embodiments of the present disclosure may also be referred to as a 5G system or an NR system, but those skilled in the art can understand its meaning. The technical solutions described in the embodiments of the present disclosure may be applicable to the 5G NR system, and may also be applicable to the subsequent evolution system of the 5G NR system.

Please refer to FIG. 2, which shows a flow chart of signal transmission in a wireless communication system. As shown in FIG. 2 , in the wireless communication system, the transmitter device 201 transmits the source information 200 through processing methods such as coding, modulation, encryption, etc., and then transmits the transmission information 202 corresponding to the source information 200 to the outside through the transmission antenna. In free space, after the transmitted information propagates in free space, it is affected by the channel environment and interference noise in free space, received by the receiving antenna, and converted into a received signal 203 that can be recognized by the receiver and transmitted to the receiver device 204. After the receiver device 204 performs data processing methods such as decoding, demodulation, and decryption on the received signal 203, it is restored to the source information or the restored information 205, and the restored information can be restored to the source information through further data processing.

Please refer to FIG. 3 , which shows a flowchart of a method for processing received information provided by an embodiment of the present application. The method may be executed by a receiving end device, wherein the above receiving end device may be in the network architecture shown in FIG. 1 . The terminal 10 or the network side device 20. The method may include the following steps:

Step 310, performing wireless signal reception to obtain reception information.

The receiving end device is a device used to implement all or part of the functions of the receiver in the wireless communication system, that is, the receiving end device can be used to implement part or all of the functions of decoding, demodulation, and decryption.

In a possible implementation manner, the received information is digital signal information formed by digital-to-analog conversion of an analog signal formed after a receiving antenna receives an electromagnetic wave signal in a free space.

In one possible implementation, the received information is a quasi-modulation symbol.

In a wireless communication system, the transmitter modulates and encodes the source information to form modulation symbols, and the modulation symbols are sent to the space channel in the free space through the antenna, and then transmitted to the receiving end device. The received signal received by the receiving end device is the quasi-modulation symbol after the modulation symbol is transmitted through the spatial channel and subjected to noise interference, and the receiving end device receives through the antenna and undergoes analog-to-digital conversion.

Step 320: Obtain transmission mode information corresponding to the received information.

In a possible implementation manner, the transmission mode information is used to indicate a transmission scheme corresponding to the received information. The transmission scheme is a configuration scheme used by the sender device to transmit the received information.

Step 330: Determine a receiver model according to the transmission mode information.

Wherein, for different transmission mode information, the receiving end device may have different receiver models, so as to realize the processing of the received information of the different transmission mode information according to the different receiver models.

Step 340: Process the received information through the receiver model to obtain the first information; the receiver model is a machine learning model obtained after training according to the received information samples and the information processing samples.

Wherein, the receiver model may be obtained by training the received information samples as input and the information processing samples as labels.

The received information sample is the same type of information as the received information, and the information processing sample is the same type of information as the first information.

To sum up, in the solution shown in the embodiment of the present application, the receiving end device uses the machine learning model trained by the training samples as the receiver model, and processes the received information according to the receiver model, and obtains the first information. Through the above scheme, the neural network model is used in the processing method of received information, and according to the different transmission methods, the receiver model used to process the received information is also different, that is, for different received information, according to different received information The transmission method uses different receiver models for data processing, which improves the accuracy of data processing.

Please refer to FIG. 4 , which shows a flowchart of a method for processing received information provided by an embodiment of the present application. The method may be executed by a receiving end device, wherein the above receiving end device may be in the network architecture shown in FIG. 1 . The terminal 10 or the network side device 20. The method may include the following steps:

Step 401, performing wireless signal reception to obtain reception information.

In a possible implementation manner, when the receiving end device is a terminal, the received information may be information formed by channel interference and device noise of the receiving end device after information delivered by the network side device.

In another possible implementation manner, when the receiving end device is a network side device, the received information may be information formed by the information uploaded by the terminal after channel interference and device noise of the network side receiving end device.

Step 402: Obtain transmission mode information corresponding to the received information.

In a possible implementation manner, the transmission mode information includes at least one of a modulation mode and multiple-in-multiple-out MIMO (Multiple-In Multiple Out, multiple-in-multiple-out) configuration information.

Among them, the modulation mode is the modulation mode corresponding to the received information, that is, after the transmitting end device corresponding to the received information modulates the source information to be sent according to the modulation mode, and then passes through the channel interference and the device noise of the receiving end device. , the received information can be formed. The MIMO configuration information is the antenna port and precoding scheme information used by the transmitting end device corresponding to the received information when sending the information corresponding to the received information.

In a possible implementation manner, when the receiving end device is a network side device, the transmission mode corresponding to the received information is determined; the transmission mode is the transmission mode used by the terminal to transmit the information corresponding to the received information; The transmission mode; according to the transmission mode, obtain the transmission mode information corresponding to the received information.

When the receiving end device is a network side device, that is, the network side device needs to receive the received information and process the received information, at this time, the network side device instructs the communication configuration between the terminal and the network side device in advance, that is, first The transmission mode between the network side device and the terminal is determined, and then the transmission mode is delivered to the terminal, so as to instruct the terminal to transmit the transmission information obtained by encoding the source information according to the transmission mode to the spatial channel.

At this time, the network side device can receive the received information formed after the transmitted information is interfered by the spatial channel, etc., and at the same time determine the transmission mode information corresponding to the received information according to the transmission mode corresponding to the received information.

For example, the above-mentioned interaction process between the network-side device and the terminal may be performed in the following manner: the network-side device first specifies the transmission mode for sending information corresponding to the terminal, and sends it to the terminal through downlink signaling, such as through broadcast messages or RRC ( Radio Resource Control) message is sent to the terminal. After receiving the downlink signaling, the terminal configures the transmission mode with the network side device according to the configuration information in the downlink signaling, and uses the transmission mode to communicate with the device. The source information is processed, and the transmitted information is sent to the free space; the network side device receives the received information formed by the interference of the transmitted information, and determines the transmission mode information corresponding to the transmission mode according to the transmission mode of the received information ( For example, the identification information corresponding to the transmission mode).

In another possible implementation manner, when the receiving end device is a terminal, the transmission mode information sent by the network side device is received.

Step 403: Determine a receiver model according to the transmission mode information.

In a possible implementation manner, the receiver model is stored in the receiver device, and after the transmission mode information is determined, the receiver model is directly selected in the receiver device according to the transmission mode information.

In another possible implementation manner, the receiver model is stored in the corresponding sender device. After the transmission mode information is determined, the receiver device can send a model acquisition request to the sender device, so that the sender device can download the model. Post this model.

In a possible implementation manner, the model structure and/or model parameters of the receiver model are determined according to the transmission mode information.

After the transmission mode information of the received information amount is determined, the model structure of the receiver model can be determined according to the transmission mode information; or the model parameters of the receiver model can be determined according to the transmission mode; or the receiver model can be determined according to the transmission mode The model structure and model parameters of the machine model.

In a possible implementation manner, the transmission characteristic information corresponding to the received information is acquired; the receiver model is determined according to the transmission mode information and the transmission characteristic information.

In a possible implementation manner, the transmission characteristic information includes at least one of data characteristic information of the received information and channel information corresponding to the received information; the data characteristic information is used to indicate the data type of the received information; the The channel information is used to indicate the channel condition of the channel used to transmit the received information.

In a possible implementation manner, when the receiving end device is a terminal, the transmission characteristic information is determined by the terminal according to the source information corresponding to the received information and the transmission channel corresponding to the received information, for example, the data characteristic information It can be determined according to the source information corresponding to the received information, and the channel information can be determined according to the transmission channel status corresponding to the received information detected by the terminal.

In a possible implementation manner, the channel condition may be used to indicate the attenuation condition of the signal in the transmission channel corresponding to the received information. For example, the channel information may be channel state information CSI (Channel State Information), that is, the channel information may be a channel attribute of a communication link, at this time, the channel state of the channel may indicate the attenuation of the signal on each transmission path of the channel factors (i.e. signal scattering, environmental attenuation, distance attenuation, etc.).

In a possible implementation manner, the data feature information may be used to indicate at least one of the data size, data priority, and data type of the source information corresponding to the received information.

The priority of the data is used to indicate the importance of the source information corresponding to the received information; the data type is used to indicate the type information corresponding to the source information, for example, the data type is used to indicate the source information corresponding to the received information. Is configuration information or request information.

The transmission feature information corresponding to the received information may be issued through downlink signaling.

In another possible implementation manner, when the receiving end device is a network-side device, the transmission feature information may be uploaded to the network-side device after being determined by the terminal.

Step 404: Process the received information by using the receiver model to obtain first information.

In a possible implementation manner, the first information is source information corresponding to the received information; or, the first information is used to restore the source information corresponding to the received information.

Wherein, the receiver model can be used to realize all or part of the functions of the receiver. When the receiver model is used to realize all the functions of the receiver, that is, the receiver model can realize the decoding, decryption, and decoding of the received information. When all operations such as tuning are performed, the first information obtained by processing the received information through the receiver model is the source information corresponding to the received information. At this time, the receiver model can be equivalently used as a receiver to perform processing on the received signal. deal with.

When the receiver model is used to realize part of the functions of the receiver, the receiver model can realize part of the operations of decoding, decryption and demodulation of the received information, that is, after processing the received information through the receiver model The obtained first information needs to be further processed to restore the source information corresponding to the received information. At this time, the receiver model can be equivalent to a part of the receiver module for processing the received signal.

In a possible implementation manner, the receiver model is a fully connected neural network model composed of N layers of fully connected layers, N≥1, and N is an integer.

The receiver model may use a fully connected network, please refer to FIG. 5 , which shows a schematic diagram of a fully connected network involved in the embodiment of the present application. As shown in Figure 5, the fully connected network constitutes an AI (Artificial Intelligence, artificial intelligence) receiver. The fully connected network here is composed of N fully connected layers 502, and the number of neurons in each fully connected layer can be Cn. The AI receiver constituted by a fully connected network can be used to process the received information 501 as the first information 503 to realize all or part of the functions of the receiver.

In a possible implementation manner, the receiver model is a convolutional neural network model composed of M layers of convolutional layers, M ≥ 1, and M is an integer.

The receiver model may also use a convolutional neural network, and FIG. 6 shows a schematic diagram of a convolutional neural network involved in the embodiment of the present application. As shown in Figure 6, the convolutional neural network constitutes the AI receiver. The convolutional neural network here is composed of M layers of convolution layers 602, the number of convolution kernels of the mth layer of convolution layers is Km, the dimension of the convolution kernels of the mth layer is Pm, and the size of the convolution kernels is Dm=[D_1 (m), D_2(m), ... D_Pm(m)]. The AI receiver formed by the convolutional neural network can also be used to process the received information 601 into the first information 603 to realize all or part of the functions of the receiver.

Among them, the above-mentioned fully-connected network-based and convolution-based neural network designs can be used alone or in combination, and separate network layers such as activation layer, normalization layer, and quantization layer can also be added.

In a possible implementation, the receiver model includes a first fully connected layer, a second fully connected layer, A common layers, a first convolutional layer, and a second convolutional layer; each common layer includes sequential The connected third convolutional layer, normalization layer and activation layer; the A common layers are connected in sequence; A≥1, and A is an integer;

The received information is processed according to the first fully connected layer to obtain the first feature information; the first feature information is processed according to the A common layers, the first convolutional layer and the second convolutional layer to obtain the second feature information. feature information; obtain third feature information according to the first feature information and the second feature information; and obtain the first information by processing the third feature information according to the second fully connected layer.

Wherein, the above-mentioned first fully connected layer includes a single fully connected layer, or is formed by connecting at least two fully connected layers in sequence, that is to say, the number of layers of the first fully connected layer may be 1 layer, or may be 2 layers or 2 floors or more.

Similarly, the above-mentioned first convolutional layer includes a single convolutional layer, or is formed by connecting at least two convolutional layers in sequence; the second convolutional layer includes a single convolutional layer, or is sequentially connected by at least two convolutional layers The third convolutional layer includes a single convolutional layer, or is formed by connecting at least two convolutional layers in sequence; the second fully-connected layer includes a single fully-connected layer, or is formed by connecting at least one fully-connected layer in sequence.

In a possible implementation manner, the received information is input into the first fully connected layer to obtain first feature extraction information; the first feature extraction information is dimensionally adjusted to obtain the first feature information.

Among them, since the received information is continuous data in the time domain, long vector data is obtained after feature extraction between fully connected layers, which is not conducive to processing by the neural network model. Therefore, the dimension of the vector data can be adjusted as Matrix data, which is convenient for the neural network model to process the data.

In a possible implementation manner, dimension adjustment is performed on the third feature information to obtain dimension adjustment information; according to the second fully connected layer, the dimension adjustment information is processed to obtain the first information.

After the one-dimensional vector data is processed into matrix data through dimension transformation, and the second feature information is processed through the convolution layer and the common layer, the obtained data is still two-dimensional matrix data, and the data is processed at the receiver. In the process of , the final need is to obtain continuous one-dimensional data in the time domain, so the third feature information in the form of a matrix needs to be dimensionally adjusted to obtain continuous one-dimensional vector data in the time domain.

In a possible implementation manner, the first feature information is sampled to obtain the first feature sampling information; and the third feature information is obtained according to the first feature sampling information and the second feature information.

After the first feature information is processed according to the convolution layer and the common layer, more original information may be lost. In order to ensure the authenticity of the data, it is necessary to sample the first feature information to obtain the first feature sampling information. And superimpose with the second feature information (that is, the skip link in the neural network training) to form a deep residual network (ResNet) to ensure the accuracy of the neural network model decoding. Among them, the deep residual network ResNet is a residual block composed of a series of residual blocks, which can be expressed as:

x _l+1 = x _l +f(x _l , w _l )

The residual block is divided into two parts, the direct mapping part and the residual part. x _l is the direct mapping part, and f(x _l , w _l ) is the residual part, which is generally composed of two or three convolution operations. The 1-layer network must contain more image information than the 1-layer, and solve the problem of gradient disappearance when the neural network level is large.

As shown in FIG. 7 , it shows a receiver neural network model involved in the embodiment of the present application. As shown in FIG. 7 , the received signal 701 first performs feature extraction through the fully connected layer 702, and then undergoes dimension adjustment through the dimension adjustment module 703. The dimension adjusted features are sampled to obtain sampled features, and the dimension adjusted features enter the convolution synchronously. Layer 704 performs feature extraction, and then enters A serially connected common layer module 705, and then passes through the convolution layer 706 and superimposes the sampled features obtained after sampling, and then passes through the dimension adjustment module 707 after dimension adjustment, through the fully connected layer. 708 is processed, and finally the first information 709 is obtained as output. Each common layer module 705 in FIG. 7 is formed by concatenating a convolution layer, a normalization layer, and an activation function in series.

As shown in FIG. 8 , it shows a receiver neural network model involved in the embodiment of the present application. As shown in FIG. 8 , the received signal 801 first performs feature extraction through a fully connected layer composed of 1024 neurons. The received signal 801 may be 512 modulation symbols modulated by QPSK (Quadrature Phase Shift Keying, quadrature phase shift keying). The 1024 neurons in the fully-connected layer 802 perform feature extraction on the 512 modulation symbols in a fully-connected manner, and the obtained feature has a dimension of 1024*1.

The feature with the dimension of 1024*1 is then input into the leakrelu layer 803, that is, the nonlinear transformation is realized through the leakrelu activation function, and then the dimension is adjusted (adjusted to 2-dimensional data of 32*32) for sampling, and then enters 256 convolution kernels The convolution layer 804 is formed, and then enters 18 serially connected public layer modules 805 for feature extraction. Each public module consists of a 256-core convolution layer, a normalization layer, and an activation function leakrelu layer. The convolution layer 806 composed of a convolution kernel is superimposed with the original sample, and after dimension adjustment, a vector with a length of 1024*1 is obtained, and then a fully connected layer 807 composed of 1024 neurons and a sigmoid activation function layer 808 are used. , and finally obtain the first information 809 as output.

In a possible implementation manner, a training sample set is obtained; the training sample set includes the received information sample and the information processing sample; and the receiver model is trained according to the training sample set.

The received information sample is the same type of information as the received information, and the information processing sample is the same type of information as the first information. The receiver model is trained according to the training sample set, so that the receiver model can obtain first information corresponding to the received information after processing the input received information.

In a possible implementation manner, the first information is source information corresponding to the received information; the information processing sample is source information corresponding to the received information sample.

When the information processing sample is the source information corresponding to the received information sample, the received information can be processed according to the received information sample and the receiver model trained by the information processing sample to obtain the first information. One piece of information is source information corresponding to the received information. That is, when the information processing sample is the source information corresponding to the received information sample, the receiver model obtained according to the received information sample and the information processing sample can have all the functions of the receiver, so as to realize the direct processing of the received information. It is restored to the source information corresponding to the received information.

In another possible implementation manner, the first information is used to obtain the source information corresponding to the received information after the first processing, and the information processing sample is the second processing of the source information corresponding to the received information sample. owned.

Wherein, the first processing includes at least one of decryption operation, decoding operation, and demodulation operation; and the second processing includes at least one of encryption operation, encoding operation, and modulation operation.

In a possible implementation manner, the first processing is an inverse processing operation corresponding to the second processing.

For example, when the first processing is a decryption operation, the second processing is an inverse processing operation corresponding to the decryption operation, that is, an encryption operation; or, when the first processing includes a decoding operation and a demodulation operation, the second processing It is an inverse processing operation of the first processing, that is, the second processing includes an encoding operation and a modulation operation.

When the information processing sample is an encrypted information processing sample obtained after the source information corresponding to the received information sample has undergone second processing (taking the second processing as an encryption operation as an example), the information processing sample also needs to be encoded, Modulation and other operations form the transmitted information corresponding to the information processing sample, and the transmitted information is subjected to channel interference to form a received information sample and received by the receiver. At this time, the training goal of the receiver model is to process the received information sample into a corresponding information processing sample. Therefore, through the received information sample and the receiver model at the training location of the information processing sample, the received information is obtained after processing the received information. The first information should also be encrypted information source information, so the first information needs to go through the first processing (ie, the decoding operation corresponding to the second processing) to obtain the information source information.

In another possible implementation manner, when the receiver model performs processing according to the received information samples and the information processing samples (that is, the source information corresponding to the receiver samples), only the receiver model may be selected to be trained to implement the receiver model. Partially functional model. For example, when training the receiver model, a data processing model can be added after the receiver model (take the data processing model to implement the function of the decryption module as an example). At this time, the receiver model needs to input the data input after receiving the information sample. Then input the decryption module to perform decryption processing to obtain the prediction information corresponding to the information processing sample. At this time, the receiver model can update the receiver model through the back propagation algorithm according to the prediction information and the information processing sample. After processing the receiver model through a large number of received information samples and information processing samples, the receiver model can process the received information into encrypted source information, and the encrypted source information can be obtained after passing through the decryption module. Source information corresponding to the received information.

The decryption module in the above scheme can also be replaced with other functional modules, that is, through the received information sample and the information processing sample (the information source information corresponding to the received information sample), the receiver model can be trained to be able to realize the receiving information. Information samples implement all or part of the functionality of a machine learning model for data processing.

In another possible implementation manner, when the information processing sample is recovery information for recovering information source information, the received information can be processed according to the received information sample and the receiver model trained by the information processing sample. processing, to obtain the first information for recovering the source information.

In a possible implementation, the training sample set further includes at least one of sample transmission mode information and sample transmission feature information; the sample transmission feature information includes at least one of data feature information and channel feature information.

Determine a receiver model corresponding to the training sample set according to at least one of the sample transmission feature information and the sample transmission method information; and train the receiver model corresponding to the training sample set based on the training sample set .

The sample transmission mode information is information of the same type as the transmission mode information, and the sample transmission characteristic information is information of the same type as the sample transmission characteristic information. When training the receiver model through the training sample set, different receiver models can also be determined through the sample transmission mode information and the sample transmission feature information. Alternatively, after the receiver model is determined, the receiver model is trained according to the training sample set. That is, different receiver models can be determined according to the sample transmission mode information and sample transmission feature information, and the different receiver models are trained according to different training sample sets, and each receiver model after training corresponds to different transmission mode information. As well as different transmission feature information, selecting a trained receiver model according to the transmission mode information and the transmission feature information can improve the data processing effect of the receiver under the transmission mode corresponding to the transmission mode information and the transmission feature information.

In a possible implementation manner, the training process of the above receiver model is trained in the receiver device, and the trained receiver model is directly stored in the memory of the receiver device.

In a possible implementation manner, the receiver model delivered by the sender device is received.

That is, the training process of the above receiver model can also be performed in the sending end device. Before the sending end device communicates with the receiving end device, the receiver model can be delivered to the sending end device. The sending end device Then, the received information corresponding to the transmitting end device is processed according to the receiver model. At this time, only the device that receives the receiver model can process the information normally, which improves the confidentiality of the information.

In a possible implementation, the training process of the receiver model is as follows:

1) Initialize the receiver model

The training device corresponding to the receiver model initializes the weight parameters corresponding to the receiver model according to the model structure of the set receiver model, so as to obtain an initial receiver model that has not been trained. Wherein, the initialization process may be random assignment of each weight parameter of the receiver model, or a preset initial value may be input into the receiver model.

2) Input the received information samples into the receiver model

Obtain a training sample set corresponding to the receiver model, where there are received information samples corresponding to the receiver model in the training sample set, input the received information samples into the receiver model, and obtain the output of the receiver model corresponding to the received information samples The predicted sample value; then the predicted sample value and the information processing sample corresponding to the received information sample are input into a loss function, and the loss function value corresponding to the received information sample is obtained.

3) Update the weight parameters of the receiver model

After the loss function value corresponding to the received information sample is obtained according to the loss function, the receiver model can be gradient updated through the back propagation algorithm according to the loss function value. Wherein, the receiver model may be updated with gradients through a back-propagation algorithm according to one loss function value, or may be updated according to multiple loss function values (for example, through the sum of multiple loss function values or the mean of multiple loss function values), At the same time, the receiver model is updated by the back-propagation algorithm. Among them, the loss function can be a suitable loss function according to the type of the signal and the structure of the model, such as a cross entropy loss function, etc., and there is no limit here.

4) Get the trained receiver model

The above process is repeated until the training meets the specified conditions, and the trained model is obtained as a trained receiver model to process the received information. Wherein, the specified condition may be that the number of training times reaches a training threshold, or the specified condition may be that the accuracy of the receiver model being verified through the verification set is greater than the verification threshold.

The above-mentioned model training process can be applied to the above-mentioned models of different structures. For example, the above-mentioned fully-connected neural network model, convolutional neural network model, deep residual network, etc. can all train the network model weights through the above-mentioned model training process.

To sum up, in the solution shown in the embodiment of the present application, the receiving end device uses the machine learning model trained by the training samples as the receiver model, and processes the received received information according to the receiver model to obtain the first information. Through the above scheme, the neural network model is used in the processing method of received information, and according to the different transmission methods, the receiver model used to process the received information is also different, that is, for different received information, the received information can be processed according to different received information. The transmission method uses different receiver models for data processing, which improves the accuracy of data processing.

FIG. 9 is a schematic flowchart of a receiving end device implementing a method for processing received information according to an exemplary embodiment of the present application. The receiving end device is a terminal, and the receiving information received by the terminal is sent from the network side device, as shown in Figure 9:

S901, the terminal receives the transmission mode information delivered by the network side device.

That is, when the terminal communicates with the network-side device by receiving information, it needs to receive the transmission mode information sent by the network-side device first, so as to determine the terminal's processing method for the information sent by the network-side device.

S902, the terminal acquires reception information.

The received information is the transmitted information obtained by the network-side device after encoding, modulation, and encryption according to the source information, and is formed by the interference of the channel and the like in the free space.

S903, the terminal determines the receiver model according to the transmission mode information delivered by the network side device.

In a possible implementation manner, the receiver model may be pre-trained by the terminal and stored in the memory of the terminal.

In another possible implementation manner, the receiver model may be delivered to the terminal after training by the network-side device, and stored in the terminal-side memory.

In another possible implementation manner, the receiver model may also be delivered by the network side device to the terminal according to downlink signaling in response to the sending information corresponding to the sending source information.

S904, the terminal processes the received information according to the receiver model corresponding to the transmission mode information to obtain the first information.

The first information may be source information corresponding to the received information, or may be recovery information that needs to be processed to obtain the source information.

FIG. 10 is a schematic flowchart of a receiving end device implementing a method for processing received information according to an exemplary embodiment of the present application. The receiving end device is a network side device, and the reception information received by the network side device is sent by the terminal. As shown in Figure 10:

S1001, a network side device determines a transmission mode.

That is, when the terminal sends information corresponding to the source information to the free space, it needs to obtain the indication information of the network side device first, so as to determine the transmission mode of the source information.

S1002, the terminal receives the transmission mode delivered by the network side device.

S1003, the terminal sends the sending information corresponding to the source information to the space according to the transmission mode, and the network side device receives the receiving information corresponding to the source information.

The received information is obtained after the transmitted information undergoes interference such as channels.

S1004, the network side device determines the transmission mode information corresponding to the received information according to the transmission mode delivered to the terminal.

The transmission mode information may be an ID identifier corresponding to the transmission mode.

S1005, the network side device determines the receiver model according to the transmission mode information.

The receiver model is pre-trained by the network-side device, and the network-side device can determine the receiver model corresponding to the transmission mode information according to the transmission mode information (ID identifier).

S1006, the network-side device processes the received information by using the receiver model to obtain first information.

The following are apparatus embodiments of the present application, which can be used to execute the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

Please refer to FIG. 11 , which shows a block diagram of an apparatus for processing received information provided by an embodiment of the present application. The device is used for a receiving end device, and has the function of implementing the above-mentioned method for processing received information. As shown in Figure 11, the apparatus may include:

A wireless signal receiving module 1101, configured to perform wireless signal reception and obtain reception information;

a transmission mode information acquisition module 1102, configured to acquire transmission mode information corresponding to the received information;

a model determination module 1103, configured to determine a receiver model according to the transmission mode information;

The first information acquisition module 1104 is configured to process the received information through the receiver model to obtain the first information; the receiver model takes the received information sample as input, and uses the information processing sample as the label for training Get a machine learning model.

In a possible implementation manner, the first information is used to obtain the source information corresponding to the received information after the first processing; the information processing sample is that the source information corresponding to the received information sample is processed through the first processing. obtained after the second treatment.

In a possible implementation manner, when the receiving end device is a network side device, the apparatus further includes:

a transmission mode determination module, configured to determine the transmission mode corresponding to the received information; the transmission mode is the transmission mode used by the terminal to transmit the sent information corresponding to the received information;

a transmission mode issuing module, configured to issue the transmission mode to the terminal;

The transmission mode information acquisition module 1102 is used to:

According to the transmission mode, obtain the transmission mode information corresponding to the received information.

In a possible implementation manner, when the receiving end device is a terminal, the transmission mode information acquisition module 1102 is configured to:

Receive the transmission mode information sent by the network side device.

In a possible implementation manner, the transmission mode information includes at least one of a modulation mode and multiple-input multiple-output MIMO configuration information.

In a possible implementation manner, the module determines a module for,

determining the model structure and/or model parameters of the receiver model according to the transmission mode information;

In a possible implementation, the apparatus further includes:

a transmission characteristic information acquisition module, configured to acquire transmission characteristic information corresponding to the received information;

The model determination module 1103 is further configured to:

The receiver model is determined according to the transmission mode information and the transmission characteristic information.

In a possible implementation manner, the transmission characteristic information includes at least one of data characteristic information of the received information and channel information corresponding to the received information;

The data characteristic information is used to indicate the data type of the received information; the channel information is used to indicate the channel status of the channel used to transmit the received information.

In a possible implementation manner, the receiver model is a convolutional neural network model composed of M layers of convolutional layers, M≥1, and M is an integer.

In a possible implementation manner, the receiver model includes a first fully connected layer, a second fully connected layer, A common layers, a first convolutional layer and a second convolutional layer; each common layer contains The third convolution layer, the normalization layer and the activation layer connected in sequence; the A common layers are connected in sequence; A≥1, and A is an integer;

The first information acquisition module 1104 includes:

a first feature information obtaining unit, configured to process the received information according to the first fully connected layer to obtain first feature information;

a second feature information obtaining unit, configured to process the first feature information according to the A common layers, the first convolution layer and the second convolution layer to obtain second feature information;

a third feature information obtaining unit, configured to obtain third feature information according to the first feature information and the second feature information;

The first information obtaining unit is configured to process the third feature information according to the second fully connected layer to obtain the first information.

In a possible implementation, the first feature information acquisition unit includes:

a first feature extraction subunit, configured to input the received information into the first fully connected layer to obtain first feature extraction information;

a first dimension adjustment unit, configured to perform dimension adjustment on the first feature extraction information to obtain the first feature information;

The first information acquisition unit includes:

A second dimension adjustment subunit, configured to perform dimension adjustment on the third feature information to obtain dimension adjustment information;

The first information obtaining subunit is configured to process the dimension adjustment information according to the second fully connected layer to obtain the first information.

In a possible implementation manner, the third feature information obtaining unit is further configured to:

Sampling the first feature information to obtain first feature sampling information;

The third feature information is obtained according to the first feature sampling information and the second feature information.

In a possible implementation, the apparatus further includes:

a training sample set acquisition module, used for acquiring a training sample set; the training sample set includes the received information samples and the information processing samples;

A receiver training module, configured to train the receiver model according to the training sample set.

In a possible implementation, the training sample set further includes at least one of sample transmission mode information and sample transmission feature information; the sample transmission feature information includes at least one of data feature information and channel feature information ;

The receiver training module includes:

a receiver model determining unit, configured to determine a receiver model corresponding to the training sample set according to at least one of the sample transmission feature information and the sample transmission mode information;

A receiver training submodule, configured to train the receiver model corresponding to the training sample set based on the training sample set.

In a possible implementation, the apparatus further includes:

Module The receiving module is used to receive the receiver model sent by the sender device.

To sum up, in the solution shown in the embodiment of the present application, the receiving end device uses the machine learning model trained by the training samples as the receiver model, and processes the received information according to the receiver model, and obtains the first information. Through the above scheme, the neural network model is used in the processing method of received information, and according to the different transmission methods, the receiver model used to process the received information is also different, that is, for different received information, the received information can be processed according to different received information. The transmission method uses different receiver models for data processing, which improves the accuracy of data processing.

It should be noted that, when the device provided in the above embodiment realizes its functions, only the division of the above functional modules is used as an example for illustration. In practical applications, the above functions can be allocated to different functional modules according to actual needs. That is, the content structure of the device is divided into different functional modules to complete all or part of the functions described above.

Regarding the apparatus in the above-mentioned embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be described in detail here.

Please refer to FIG. 12 , which shows a schematic structural diagram of a computer device 1200 provided by an embodiment of the present application. The computer device 1200 may include: a processor 1201 , a receiver 1202 , a transmitter 1203 , a memory 1204 and a bus 1205 .

The processor 1201 includes one or more processing cores, and the processor 1201 executes various functional applications and information processing by running software programs and modules.

The receiver 1202 and the transmitter 1203 may be implemented as a communication component, which may be a communication chip. The communication chip may also be referred to as a transceiver.

The memory 1204 is connected to the processor 1201 through the bus 1205 .

The memory 1204 can be used to store a computer program, and the processor 1201 is used to execute the computer program to implement each step performed by the server device, configuration device, cloud platform or account server in the above method embodiments.

Additionally, memory 1204 may be implemented by any type or combination of volatile or non-volatile storage devices including, but not limited to, magnetic or optical disks, electrically erasable programmable Read Only Memory, Erasable Programmable Read Only Memory, Static Anytime Access Memory, Read Only Memory, Magnetic Memory, Flash Memory, Programmable Read Only Memory.

In an exemplary embodiment, the computer device includes a processor, a memory, and a transceiver (the transceiver may include a receiver for receiving information and a transmitter for transmitting information);

In a possible implementation, when the computer device is implemented as a receiver device, the terminal includes a processor, a memory and a transceiver;

the receiver, configured to perform wireless signal reception and obtain reception information;

the processor, configured to obtain the transmission mode information corresponding to the received information;

the processor, configured to determine a receiver model according to the transmission mode information;

The processor is configured to process the received information through the receiver model to obtain the first information; the receiver model is a machine learning model obtained after training according to the received information samples and the information processing samples.

The processor and transceiver in the receiving end device involved in the embodiment of the present application may perform the steps performed by the receiving end device in the above-mentioned received information processing method shown in FIG. 3 or FIG. 4 , which will not be repeated here.

An embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored in the storage medium, and the computer program is loaded and executed by a processor to implement the received information shown in any of the above-mentioned FIG. 3 or FIG. 4 . The various steps in the processing method.

The application also provides a computer program product or computer program, the computer program product or computer program comprising computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes each step in the method for processing received information shown in FIG. 3 or FIG. 4 .

Those skilled in the art should realize that, in one or more of the above examples, the functions described in the embodiments of the present application may be implemented by hardware, software, firmware, or any combination thereof. When implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage medium can be any available medium that can be accessed by a general purpose or special purpose computer.

The above descriptions are only exemplary embodiments of the present application, and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the protection of the present application. within the range.

Claims

A method for processing received information, wherein the method is used for a receiving end device, and the method includes:

Perform wireless signal reception to obtain reception information;

obtaining the transmission mode information corresponding to the received information;

determining a receiver model according to the transmission mode information;

The received information is processed by the receiver model to obtain the first information; the receiver model is a machine learning model obtained after training according to the received information samples and the information processing samples.
The method according to claim 1, wherein the first information is source information corresponding to the received information; and the information processing sample is source information corresponding to the received information sample.
The method according to claim 1, wherein the first information is used to obtain information source information corresponding to the received information after first processing; the information processing sample is the information corresponding to the received information sample. The source information is obtained after the second processing.
The method of claim 3, wherein the first processing is an inverse processing operation corresponding to the second processing.
The method according to claim 1, wherein when the receiving end device is a network side device, before acquiring the transmission mode information corresponding to the received information, the method further comprises:

Determine the transmission mode corresponding to the received information; the transmission mode is the transmission mode used by the terminal to transmit the sent information corresponding to the received information;

delivering the transmission mode to the terminal;

The acquiring the transmission mode information corresponding to the received information includes:

According to the transmission mode, obtain the transmission mode information corresponding to the received information.
The method according to claim 5, wherein when the receiving end device is a terminal, the acquiring the transmission mode information corresponding to the received information comprises:

Receive the transmission mode information sent by the network side device.
The method according to any one of claims 1 to 6, wherein the transmission mode information includes at least one of modulation mode and MIMO configuration information.
The method according to any one of claims 1 to 6, wherein the determining the receiver model according to the transmission mode information comprises:

According to the transmission mode information, a model structure and/or model parameters of the receiver model are determined.
The method according to any one of claims 1 to 6, wherein before determining the receiver model according to the transmission mode, the method further comprises:

obtaining transmission feature information corresponding to the received information;

The determining of the receiver model according to the transmission mode information includes:

The receiver model is determined according to the transmission mode information and the transmission characteristic information.
The method according to claim 9, wherein the transmission characteristic information comprises at least one of data characteristic information of the received information and channel information corresponding to the received information;

The data characteristic information is used to indicate the data type of the received information; the channel information is used to indicate the channel status of the channel used to transmit the received information.
The method according to any one of claims 1 to 6, wherein the receiver model is a fully connected neural network model composed of N layers of fully connected layers, N≥1, and N is an integer.
The method according to any one of claims 1 to 6, wherein the receiver model is a convolutional neural network model composed of M layers of convolutional layers, M≥1, and M is an integer.
The method according to any one of claims 1 to 6, wherein the receiver model comprises a first fully connected layer, a second fully connected layer, A common layers, a first convolutional layer and a second convolutional layer layer; each common layer includes a third convolution layer, a normalization layer and an activation layer connected in sequence; the A common layers are connected in sequence; A≥1, and A is an integer;

According to the receiver model, the received information is processed to obtain the first information, including:

Process the received information according to the first fully connected layer to obtain first feature information;

According to the A common layers, the first convolution layer and the second convolution layer, the first feature information is processed to obtain the second feature information;

obtaining third feature information according to the first feature information and the second feature information;

According to the second fully connected layer, the third feature information is processed to obtain the first information.
The method according to claim 13, wherein the processing of the received information according to the first fully-connected layer to obtain the first feature information comprises:

Inputting the received information into the first fully connected layer to obtain first feature extraction information;

Perform dimension adjustment on the first feature extraction information to obtain the first feature information;

The processing of the third feature information according to the second fully connected layer to obtain the first information includes:

performing dimension adjustment on the third feature information to obtain dimension adjustment information;

According to the second fully connected layer, the dimension adjustment information is processed to obtain the first information.
The method according to claim 13, wherein the obtaining third characteristic information according to the first characteristic information and the second characteristic information comprises:

Sampling the first feature information to obtain first feature sampling information;

The third feature information is obtained according to the first feature sampling information and the second feature information.
The method according to any one of claims 1 to 6, wherein the method further comprises:

Obtain a training sample set; the training sample set includes the received information samples and the information processing samples;

The receiver model is trained according to the training sample set.
The method according to claim 16, wherein the training sample set further includes at least one of sample transmission mode information and sample transmission characteristic information; the sample transmission characteristic information includes data characteristic information and channel characteristic information. at least one of;

The training of the receiver model according to the training sample set further includes:

determining a receiver model corresponding to the training sample set according to at least one of the sample transmission feature information and the sample transmission mode information;

Based on the training sample set, the receiver model corresponding to the training sample set is trained.
The method according to any one of claims 1 to 6, wherein before determining the receiver model according to the transmission mode information, the method further comprises:

Receives the receiver model delivered by the sender device.
An apparatus for processing received information, characterized in that the apparatus is used for receiving end equipment, and the apparatus comprises:

The wireless signal receiving module is used to perform wireless signal receiving and obtain receiving information;

a transmission mode information acquisition module, configured to acquire the transmission mode information corresponding to the received information;

a model determination module, configured to determine a receiver model according to the transmission mode information;

a first information acquisition module, configured to process the received information through the receiver model to obtain the first information; the receiver model takes the received information sample as an input, and uses the information processing sample as a label to obtain the training machine learning model.
The apparatus according to claim 19, wherein the first information is source information corresponding to the received information; and the information processing sample is source information corresponding to the received information sample.
The apparatus according to claim 19, wherein the first information is used to obtain information source information corresponding to the received information after first processing; the information processing sample is the information corresponding to the received information sample. The source information is obtained after the second processing.
The apparatus according to claim 21, wherein the first processing is an inverse processing operation corresponding to the second processing.
The apparatus according to claim 19, wherein when the receiving end device is a network side device, the apparatus further comprises:

a transmission mode determination module, configured to determine the transmission mode corresponding to the received information; the transmission mode is the transmission mode used by the terminal to transmit the sent information corresponding to the received information;

a transmission mode issuing module, configured to issue the transmission mode to the terminal;

The transmission mode information acquisition module is used for,

According to the transmission mode, the transmission mode information corresponding to the received information is acquired.
The apparatus according to claim 19, wherein when the receiving end device is a terminal, the transmission mode information acquisition module is used for:

Receive the transmission mode information sent by the network side device.
The apparatus according to any one of claims 19 to 24, wherein the transmission mode information comprises at least one of modulation mode and MIMO configuration information.
The device according to any one of claims 19 to 24, wherein the module determination module is configured to:

According to the transmission mode information, a model structure and/or model parameters of the receiver model are determined.
The device according to any one of claims 19 to 24, wherein the device further comprises:

a transmission characteristic information acquisition module, configured to acquire transmission characteristic information corresponding to the received information;

The model determination module is also used for,

The receiver model is determined according to the transmission mode information and the transmission characteristic information.
The apparatus according to claim 27, wherein the transmission characteristic information comprises at least one of data characteristic information of the received information and channel information corresponding to the received information;

The data characteristic information is used to indicate the data type of the received information; the channel information is used to indicate the channel status of the channel used to transmit the received information.
The apparatus according to any one of claims 19 to 24, wherein the receiver model is a fully connected neural network model composed of N layers of fully connected layers, N≥1, and N is an integer.
The apparatus according to any one of claims 19 to 24, wherein the receiver model is a convolutional neural network model composed of M layers of convolutional layers, where M≥1, and M is an integer.
The apparatus according to any one of claims 19 to 24, wherein the receiver model comprises a first fully connected layer, a second fully connected layer, A common layers, a first convolutional layer, and a second convolutional layer layer; each common layer includes a third convolution layer, a normalization layer and an activation layer connected in sequence; the A common layers are connected in sequence; A≥1, and A is an integer;

The first information acquisition module includes:

a first feature information obtaining unit, configured to process the received information according to the first fully connected layer to obtain first feature information;

a second feature information obtaining unit, configured to process the first feature information according to the A common layers, the first convolution layer and the second convolution layer to obtain second feature information;

a third feature information obtaining unit, configured to obtain third feature information according to the first feature information and the second feature information;

The first information obtaining unit is configured to process the third feature information according to the second fully connected layer to obtain the first information.
The device according to claim 31, wherein the first feature information acquisition unit comprises:

a first feature extraction subunit, configured to input the received information into the first fully connected layer to obtain first feature extraction information;

a first dimension adjustment unit, configured to perform dimension adjustment on the first feature extraction information to obtain the first feature information;

The first information acquisition unit includes:

A second dimension adjustment subunit, configured to perform dimension adjustment on the third feature information to obtain dimension adjustment information;

The first information obtaining subunit is configured to process the dimension adjustment information according to the second fully connected layer to obtain the first information.
The device according to claim 31, wherein the third feature information obtaining unit is further configured to sample the first feature information to obtain the first feature sampling information;

The third feature information is obtained according to the first feature sampling information and the second feature information.
The device according to any one of claims 19 to 24, wherein the device further comprises:

a training sample set acquisition module, used to obtain a training sample set; the training sample set includes the received information samples and the information processing samples;

A receiver training module, configured to train the receiver model according to the training sample set.
The apparatus according to claim 34, wherein the training sample set further includes at least one of sample transmission mode information and sample transmission characteristic information; the sample transmission characteristic information includes data characteristic information and channel characteristic information. at least one of;

The receiver training module includes:

a receiver model determining unit, configured to determine a receiver model corresponding to the training sample set according to at least one of the sample transmission feature information and the sample transmission mode information;

A receiver training submodule, configured to train the receiver model corresponding to the training sample set based on the training sample set.
The method according to any one of claims 19 to 24, wherein the device further comprises:

Module The receiving module is used to receive the receiver model sent by the sender device.
A computer device, characterized in that the computer device is a receiver device, and the receiver device includes a processor, a memory, and a transceiver;

the receiver, configured to perform wireless signal reception and obtain reception information;

the processor, configured to obtain the transmission mode information corresponding to the received information;

the processor, configured to determine a receiver model according to the transmission mode information;

The processor is configured to process the received information through the receiver model to obtain the first information; the receiver model is a machine learning model obtained after training according to the received information samples and the information processing samples.
A computer-readable storage medium, characterized in that a computer program is stored in the storage medium, and the computer program is used to be executed by a processor to realize the received information processing according to any one of claims 1 to 18 method.