Communication method and device based on network communication model, medium and storage medium

Abstract

The network communication model comprises a channel layer and a plurality of signal transmission modules, each signal transmission module comprises a signal emission module and a signal output module, and each signal emission module comprises an encoder network, a power distribution network and a coding modulation layer. The communication method comprises the following steps: encoding based on a received to-be-transmitted signal and a preset interference coefficient by using each encoder network to obtain a plurality of first encoding signals, encoding based on the preset interference coefficient by using each power distribution network to obtain a plurality of second encoding signals, and transmitting the plurality of second encoding signals to the corresponding power distribution network; transmitting signals transmitted by the signal transmitting modules are obtained based on the first coding signals and the second coding signals through the coding modulation layer, and input signals of the signal output modules are calculated based on preset noise, a preset interference coefficient and the transmitting signals through the channel layer; the signal output modules are respectively used for decoding based on the input signals to obtain a plurality of output signals, so that the parameter quantity of an encoder network is reduced, and the encoding modulation efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a communication method, device, medium and storage medium based on a network communication model. Background Art

[0002] In the prior art, among the main methods of physical layer coding modulation in the communication field, in addition to some traditional methods such as Hamming code, QPSK (Quadrature Phase Shift Keying), etc., some artificial intelligence methods are also used, such as linear neural network layers, such as multi-layer perceptrons, MLP (Multi-Layer Perceptron) to fit the characteristics of the sequence data to be transmitted. This method works well in traditional deep learning tasks, but has a series of defects in communication scenarios: first, due to the lack of data sets, linear neural networks often show low fitting efficiency when processing complex nonlinear features, resulting in reduced model accuracy; second, the complexity of network parameters requires the model to consume more computing resources during reasoning, further affecting real-time performance; finally, although the use of models such as recursive neural networks (RNN), long short-term memory networks (LSTM) and attention mechanisms can significantly improve the accuracy of sequence tasks, their reasoning speed is slow and the network parameters are often large, resulting in reduced practical value in real-time communication scenarios. Summary of the invention

[0003] In view of the deficiencies of the above-mentioned prior art, the present application proposes a communication method, device, medium and storage medium based on a network communication model.

[0004] In a first aspect, an embodiment of the present application provides a communication method based on a network communication model, wherein the network communication model includes a channel layer and multiple signal transmission modules, each of which includes a signal transmission module and a signal output module; each of the signal transmission modules includes an encoder network, a power distribution network and a coding modulation layer, and the communication method includes: using each of the encoder networks to encode based on the received signal to be transmitted and a preset interference coefficient to obtain multiple first coded signals; using each of the power distribution networks to encode based on the preset interference coefficient to obtain multiple second coded signals; using the coding modulation layer to obtain the first coded signal encoded by the encoder network in the signal transmission module corresponding to each of the signal transmission modules, and the second coded signal encoded by the power distribution network, to obtain the transmission signal transmitted by the signal transmission module corresponding to each of the signal transmission modules; using the channel layer to calculate the input signal of the signal output module corresponding to each of the signal transmission modules based on the preset noise, the preset interference coefficient and each of the transmission signals; using the signal output modules corresponding to each of the signal transmission modules to decode based on each of the input signals to obtain multiple output signals.

[0005] In a second aspect, an embodiment of the present application provides a communication device of a network communication model, comprising: the network communication model comprises a channel layer and a plurality of signal transmission modules, each of the signal transmission modules comprises a signal transmission module and a signal output module; each of the signal transmission modules comprises an encoder network, a power distribution network and a coding modulation layer, and the communication device comprises: a first encoding module, for respectively using each of the encoder networks, encoding based on the received signal to be transmitted and a preset interference coefficient, to obtain a plurality of first encoded signals; a second encoding module, for respectively using each of the power distribution networks, encoding based on the preset interference coefficient, to obtain a plurality of second encoded signals; a coding modulation module, for The coding modulation layer is used to obtain the first coded signal obtained by the encoder network coding in the signal transmission module corresponding to each signal transmission module, and the second coded signal obtained by the power allocation network coding, to obtain the transmission signal transmitted by the signal transmission module corresponding to each signal transmission module; the calculation module is used to use the channel layer to calculate the input signal of the signal output module corresponding to each signal transmission module based on the preset noise, the preset interference coefficient and each transmission signal; the decoding module is used to use the signal output module corresponding to each signal transmission module to decode based on each input signal to obtain multiple output signals.

[0006] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method described in the first aspect above is implemented.

[0007] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect above is implemented.

[0008] The technical solution provided in the embodiments of the present application has at least the following technical effects or advantages:

[0009] The network communication model of the embodiment of the present application includes a channel layer and multiple signal transmission modules, each signal transmission module includes a signal transmission module and a signal output module, each signal transmission module includes an encoder network, a power distribution network and a coding modulation layer, each encoder network is used to encode based on the received signal to be transmitted and the preset interference coefficient to obtain multiple first coded signals, each power distribution network is used to encode based on the preset interference coefficient to obtain multiple second coded signals, and then the coding modulation layer is used to obtain the first coded signal encoded by the encoder network in the signal transmission module corresponding to each signal transmission module and the second coded signal encoded by the power distribution network to obtain the transmission signal transmitted by the signal transmission module corresponding to each signal transmission module, and the channel layer is used to calculate the input signal of the signal output module corresponding to each signal transmission module based on the preset noise, the preset interference coefficient and each transmission signal, and finally, the signal output module corresponding to each signal transmission module is used to decode based on each input signal to obtain multiple output signals, so that the number of parameters required for the network communication model is greatly reduced, a lower bit error rate is achieved, the computing power requirement is reduced, and the practicality of the deep learning network in physical layer communication is improved.

[0010] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0012] Figure 1 A schematic diagram showing the structure of a network communication model provided in an embodiment of the present application is shown;

[0013] Figure 2 A flow chart of a communication method based on a network communication model provided in an embodiment of the present application is shown;

[0014] Figure 3 A schematic diagram of the structure of a communication device based on a network communication model provided in an embodiment of the present application is shown;

[0015] Figure 4 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0016] The exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0017] In the prior art, among the main methods of physical layer coding modulation in the communication field, in addition to some traditional methods such as Hamming code, QPSK (Quadrature Phase Shift Keying), etc., some artificial intelligence methods are also used, such as linear neural network layers, such as multi-layer perceptrons, MLP (Multi-Layer Perceptron) to fit the characteristics of the sequence data to be transmitted. This method works well in traditional deep learning tasks, but has a series of defects in communication scenarios: first, due to the lack of data sets, linear neural networks often show low fitting efficiency when processing complex nonlinear features, resulting in reduced model accuracy; second, the complexity of network parameters requires the model to consume more computing resources during reasoning, further affecting real-time performance; finally, although the use of models such as recursive neural networks (RNN), long short-term memory networks (LSTM) and attention mechanisms can significantly improve the accuracy of sequence tasks, their reasoning speed is slow and the network parameters are often large, resulting in reduced practical value in real-time communication scenarios.

[0018] Figure 1 A structural schematic diagram of a network communication model on which the communication method provided in an embodiment of the present application is based, the network communication model comprising a channel layer and multiple signal transmission modules, each signal transmission module comprising a signal transmission module and a signal output module, the signal transmission module being connected to the channel layer and the signal output module in sequence, each signal transmission module comprising an encoder network, a power distribution network and a coding modulation layer, the encoder network and the power distribution network being connected to the coding modulation layer respectively.

[0019] The first encoder network includes a first input layer, a first encoding layer, a connection layer, a second encoding layer and a normalization layer, wherein the first input layer, the first encoding layer, the first connection layer, the second encoding layer and the normalization layer are connected in sequence, the first input layer includes multiple input nodes, the first encoding layer includes multiple activation functions, each input node is connected to each activation function respectively, the second encoding layer has the same structure as the first encoding layer, the first connection layer includes multiple addition nodes, each activation function of the first encoding layer is connected to multiple addition nodes in a one-to-one correspondence, and then the multiple addition nodes are also connected to each activation function of the second encoding layer in a one-to-one correspondence, wherein the first encoder network is any one of the multiple encoder networks.

[0020] The structures of each power distribution network are similar to those of each encoder network. The difference is that the activation function of the encoding layer in each power distribution network is smaller than the activation function of the encoding layer in the encoder network. The specific structure is shown in the figure and will not be repeated here.

[0021] The channel layer includes a plurality of noise adding layers and a plurality of noise layers. Each signal transmitting module is connected to each noise layer in a one-to-one correspondence, and a noise adding layer is connected between two adjacent noise layers.

[0022] Each signal output module includes a decoder network, and the structure of the decoder network includes a second normalization layer, a first decoding layer, a second connection layer, a second decoding layer and an output layer, wherein the second normalization layer has the same structure as the normalization layer in the first encoder network, the first decoding layer is similar to the structure of the first encoding layer in the first encoder network, the first decoding layer includes multiple Sigmoid activation functions, similarly, the second decoding layer has the same structure as the first decoding layer, the output layer has the same structure as the first input layer in the first encoder network, and the connections of the second normalization layer, the third encoding layer, the second connection layer, the fourth encoding layer and the output layer in the decoder network are opposite to those in the first encoder network, and will not be repeated here.

[0023] The network communication model of the embodiment of the present application can support communication scenarios of resource-constrained physical layer transceivers, reduce dependence on large amounts of data and computing resources, and by integrating knowledge in the field of communications into network design, the model-driven approach can better solve problems related to communication signal processing, thereby improving communication performance; moreover, the network communication model of the embodiment of the present application can also support coding and modulation designs with low bit error rate requirements, and by adjusting the communication scheme according to changes in channel conditions, the bit error rate performance of communications can be improved, ensuring high-speed transmission and reliability under various channel conditions.

[0024] Figure 2 A communication method based on a network communication model is provided for the embodiment of the present application. The scheme of the embodiment of the present application is specifically described below in conjunction with the accompanying drawings.

[0025] See also Figure 2 A flow chart of a communication method based on a network communication model is shown, where the network model is the network communication model described in the above embodiment, and the method specifically comprises the following steps:

[0026] Step 101: using each encoder network respectively, encoding is performed based on a received signal to be transmitted and a preset interference coefficient to obtain a plurality of first encoded signals.

[0027] In one embodiment, the first encoder network includes a first input layer, a first encoding layer, a first connection layer, a second encoding layer and a normalization layer, wherein the first encoder network is any one of a plurality of encoder networks. Then, the first input layer can be used to align the received signal to be transmitted and the preset interference coefficient, and the first encoding layer can be used to encode the aligned signal to be transmitted to obtain a first intermediate coded signal, the first intermediate coded signal includes a plurality of first activation values, and the first connection layer is used to connect the first activation values ​​to obtain a second intermediate transmission signal, and the second encoding layer is used to encode the second intermediate transmission signal to obtain a third intermediate transmission signal, and the normalization layer is used to perform normalization based on the third intermediate transmission signal to obtain the first coded signal. Finally, based on the preset interference coefficient, the target signal output by the target input layer is channel equalized, and the target signal is input to the next target input layer adjacent to the target input layer, wherein the target input layer includes any layer among the input layer, the first encoding layer, the first connection layer, the second encoding layer and the normalization layer.

[0028] On the basis of the above implementation, in some modified factual modes, the first input layer includes multiple input nodes, the first coding layer includes multiple activation functions, and the first coding layer is used to encode the signal to be transmitted after the alignment processing to obtain the first intermediate coded signal, which may specifically include: using the first activation function to calculate the first activation value based on the signal to be transmitted after the alignment processing transmitted by each input node, the first activation function is any one of the multiple activation functions, wherein the activation function is shown in the following formula (1):

[0029]

[0030] Among them, φ(x j ) represents the first activation value, w b represents the first learning parameter, w s represents the second learning parameter, b(x j ) represents the basic activation function, B i (x j ) represents the spline function, c i represents the coefficient of the spline function, i represents the number of spline functions, x jRepresents the signal to be transmitted after the alignment processing transmitted by the j-th input node.

[0031] On the basis of the above implementation, in some modified implementations, the first connection layer includes a plurality of adding nodes, each adding node respectively adds the activation values ​​calculated by each activation function to obtain each second intermediate transmission signal.

[0032] Furthermore, the second coding layer also includes a plurality of activation functions, which are the same as the activation functions of the first coding layer. The second activation values ​​are calculated based on the second intermediate transmission signals using the activation functions. In this case, φ(x j ) represents the second activation value, w b represents the first learning parameter, w s represents the second learning parameter, b(x j ) represents the basic activation function, B i (x j ) represents the spline function, c i represents the coefficient of the spline function, i represents the number of spline functions, x j represents the j-th second intermediate transmission signal, and uses the second activation value as the third intermediate transmission signal.

[0033] Furthermore, a normalization layer is used to perform normalization processing based on each third intermediate transmission signal to obtain a first coded signal.

[0034] Step 102: Using each power allocation network respectively, encoding is performed based on a preset interference coefficient to obtain a plurality of second encoded signals.

[0035] In the embodiment of the present application, the encoding method of each power distribution network is the same as that of the encoder network, which will not be repeated here.

[0036] Step 103: Using the coding modulation layer, the first coded signal obtained by the encoder network coding in the signal transmission module corresponding to each signal transmission module and the second coded signal obtained by the power allocation network coding are obtained to obtain the transmission signal transmitted by the signal transmission module corresponding to each signal transmission module.

[0037] In one embodiment, a first coding modulation layer is used to multiply a first coded signal obtained by encoder network coding in a signal transmission module corresponding to a first signal transmission module with a second coded signal obtained by power allocation network coding in a signal transmission module corresponding to the first signal transmission module to obtain a transmission signal transmitted by the signal transmission module corresponding to the first signal transmission module, wherein the first coding modulation layer is a coding modulation layer in a signal transmission module corresponding to the first signal transmission module, and the first signal transmission module is any one of a plurality of signal transmission modules.

[0038] Step 104: Calculate the input signal of the signal output module corresponding to each signal transmission module based on the preset noise, the preset interference coefficient and each transmission signal by using the channel layer.

[0039] In one implementation, the channel layer is used to calculate the input signal of the signal output module corresponding to each signal transmission module based on the preset noise, the preset interference coefficient and each transmission signal by the following formula (2):

[0040]

[0041] in, represents the input signal of the output module corresponding to the mth signal transmission module, represents the transmission signal transmitted by the signal transmission module corresponding to the mth signal transmission module, α represents the preset interference coefficient, represents the transmission signal transmitted by the signal transmission module corresponding to the nth signal transmission module that interferes with the mth signal transmission module, N represents the number of signal transmission modules that interfere with the mth signal transmission module, Represents the preset noise of the mth signal transmission module.

[0042] Step 105: Decoding each input signal using the signal output module corresponding to each signal transmission module to obtain multiple output signals.

[0043] In one implementation, the input signal can be normalized by using the normalization layer in the first signal output module to eliminate the influence of the channel layer on the amplitude range of the received and transmitted signals, and multiple first intermediate decoded signals are obtained. Then, multiple Sigmond activation functions of the first decoding layer in the first signal output module are used to perform decoding processing based on each first intermediate decoded signal to obtain each second intermediate decoded signal. Multiple Sigmond activation functions of the second decoding layer in the first signal output module are used to perform decoding processing based on each second intermediate decoded signal to obtain multiple third intermediate decoded signals, which are then output by each output node of the output layer in the first signal output module to obtain multiple output signals. The first signal output module is any one of the multiple signal output modules.

[0044] On the basis of the above implementation manner, in some changed implementation manners, the first encoder network includes a first encoding layer, and the first encoder network is any one of a plurality of encoder networks. Before encoding based on the received signal to be transmitted and the preset interference coefficient using each encoder network respectively, it also includes: initializing the network communication model to be trained based on preset parameters, and the parameters include: a first number of transmission signal data sets to be trained, a second number of transmission signals to be trained processed by the network communication model to be trained, a third number of channels, the number of times the network communication model is updated, the number of activation functions in the first encoding layer, the order and grid accuracy of the spline function in the activation function, and the learning rate of the network communication model to be trained. Based on the third number, the network communication model to be trained is cyclically trained, and if the network communication model reaches the convergence condition, the network communication model is obtained.

[0045] Based on the above implementation, in some modified implementations, the convergence condition can be the number of training times for the network communication model to be trained, that is, the third number. When the number of training times reaches the third number, it is determined that the network communication model has reached the convergence condition and the network communication model is obtained.

[0046] On the basis of the above implementation manner, in some modified implementation manners, the network communication model to be trained is trained, including: randomly generating an interference coefficient to be trained and a transmission signal data set to be trained, and then, based on a predefined search space, generating noise to be trained, freezing the channel gradient, training the network communication model to be trained based on the interference coefficient to be trained, the transmission signal data set to be trained and the noise to be trained, obtaining an output signal data set to be trained, calculating a loss value based on the transmission signal data set to be trained and the output signal data set to be trained, and finally, based on the loss value, updating the weight of the network communication model to be trained using back propagation technology.

[0047] On the basis of the above implementation, in some modified implementations, the loss value is calculated based on the transmission signal data set to be trained and the output signal data set to be trained, including: calculating the loss value through a loss function based on the transmission signal data set to be trained and the output signal data set to be trained, and the loss function is as follows: Formula (3):

[0048]

[0049]

[0050] in, Represents the loss value, L i represents the first training temporary value, G represents the number of transmission signal data sets to be trained or output signal data sets to be trained, represents the i-th transmission signal to be trained, Represents the i-th output signal to be trained.

[0051] In the embodiment of the present application, each encoder network is used to encode based on the received signal to be transmitted and the preset interference coefficient to obtain multiple first encoded signals, and each power distribution network is used to encode based on the preset interference coefficient to obtain multiple second encoded signals, and then the coding modulation layer is used to obtain the first encoded signal encoded by the encoder network in the signal transmission module corresponding to each signal transmission module, and the second encoded signal encoded by the power distribution network to obtain the transmission signal transmitted by the signal transmission module corresponding to each signal transmission module, and the channel layer is used to calculate the input signal of the signal output module corresponding to each signal transmission module based on the preset noise, the preset interference coefficient and each transmission signal, and finally, the signal output module corresponding to each signal transmission module is used to decode based on each input signal to obtain multiple output signals, so that the number of parameters required for the network communication model is greatly reduced, a lower bit error rate is achieved, the computing power requirement is reduced, and the practicality of the deep learning network in the physical layer communication is improved. .

[0052] See also Figure 3 The embodiment of the present application also provides a communication device of a network communication model, which is used to execute the communication method of the network communication model described in the above embodiment. The network communication model includes a channel layer and multiple signal transmission modules, each of which includes a signal transmission module and a signal output module; each of which includes an encoder network, a power distribution network and a coding modulation layer, and the communication device includes:

[0053] A first encoding module 201, configured to respectively use the encoder networks to encode based on the received signal to be transmitted and a preset interference coefficient to obtain a plurality of first encoded signals;

[0054] A second encoding module 202, configured to respectively use each of the power distribution networks to perform encoding based on the preset interference coefficient to obtain a plurality of second encoded signals;

[0055] The coding and modulation module 203 is used to obtain the transmission signal transmitted by the signal transmission module corresponding to each of the signal transmission modules by respectively using the coding and modulation layer based on the first coded signal obtained by the encoder network coding in the signal transmission module corresponding to each of the signal transmission modules, and the second coded signal obtained by the power allocation network coding;

[0056] A calculation module 204, configured to calculate an input signal of the signal output module corresponding to each of the signal transmission modules based on a preset noise, the preset interference coefficient and each of the transmission signals by using the channel layer;

[0057] The decoding module 205 is used to respectively use the signal output modules corresponding to the signal transmission modules to decode the input signals to obtain multiple output signals.

[0058] The communication device of the network communication model provided in the embodiment of the present application and the communication method of the network communication model provided in the above embodiment are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented therein.

[0059] The present application also provides an electronic device corresponding to the communication method of the network communication model provided in the above embodiment. Figure 4 , which shows a schematic diagram of an electronic device provided by some embodiments of the present application. Figure 4 As shown, the electronic device 30 may include: a processor 300, a memory 301, a bus 302 and a communication interface 303, wherein the processor 300, the communication interface 303 and the memory 301 are connected via the bus 302; the memory 301 stores a computer program that can be run on the processor 300, and when the processor 300 runs the computer program, the communication method of the network communication model provided in any of the aforementioned embodiments of the present application is executed.

[0060] The memory 301 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one physical port 303 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0061] The bus 302 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 301 is used to store programs, and the processor 300 executes the programs after receiving the execution instruction. The communication method of the network communication model disclosed in any implementation of the embodiment of the present application may be applied to the processor 300, or implemented by the processor 300.

[0062] The processor 300 may be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor 300 or an instruction in the form of software. The above processor 300 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a readily available programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module may be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 301, and the processor 300 reads the information in the memory 301 and completes the steps of the above method in combination with its hardware.

[0063] The electronic device provided in the embodiment of the present application and the communication method of the network communication model provided in the embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented therein.

[0064] An embodiment of the present application also provides a computer-readable storage medium corresponding to the communication method of the network communication model provided in the aforementioned embodiment, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it will execute the communication method of the network communication model provided in any of the aforementioned embodiments.

[0065] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0066] An embodiment of the present application also provides a computer program product corresponding to the communication method of the network communication model provided in the aforementioned embodiment, including a computer program, which is executed by a processor to implement the communication method of the network communication model provided in the aforementioned embodiments.

[0067] The computer-readable storage medium and computer program product provided in the above-mentioned embodiments of the present application are based on the same inventive concept as the communication method of the network communication model provided in the embodiments of the present application, and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0068] It should be noted that:

[0069] The algorithm and display provided herein are not inherently related to any particular computer, virtual device or other equipment. Various general purpose devices can also be used together with the teachings based on this. According to the above description, it is obvious to construct the structure required for this type of device. In addition, the application is not directed to any specific programming language. It should be understood that the content of the application described herein can be realized using various programming languages, and the description of the specific language above is to disclose the best mode of implementation of the application.

[0070] In the description provided herein, a large number of specific details are described. However, it is understood that the embodiments of the present application can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0071] Similarly, it should be understood that in order to streamline the present application and help understand one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present application, the various features of the present application are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be interpreted as reflecting the following intention: the claimed application requires more features than the features clearly stated in each claim. More specifically, as reflected in the claims below, the inventive aspects are less than all the features of the single embodiment disclosed above. Therefore, the claims following the specific embodiment are hereby expressly incorporated into the specific embodiment, wherein each claim itself serves as a separate embodiment of the present application.

[0072] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0073] In addition, those skilled in the art will appreciate that, although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present application and form different embodiments. For example, in the claims below, any one of the claimed embodiments may be used in any combination.

[0074] The various component embodiments of the present application can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) can be used in practice to implement some or all functions of some or all components in the creation device of the virtual machine according to the embodiment of the present application. The application can also be implemented as a device or device program (e.g., computer program and computer program product) for executing part or all of the methods described herein. Such a program implementing the present application can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0075] It should be noted that the above embodiments illustrate the present application rather than limit the present application, and that those skilled in the art may design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets should not be constructed as a limitation to the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of multiple such elements. The present application may be implemented by means of hardware including several different elements and by means of a suitably programmed computer. In a unit claim that lists several devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names.

[0076] The above is only a preferred specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A communication method based on a network communication model, characterized in that: The network communication model includes a channel layer and a plurality of signal transmission modules, each of which includes a signal transmission module and a signal output module; each of which includes an encoder network, a power distribution network and a coding modulation layer, and the communication method includes: Using each of the encoder networks respectively, encoding is performed based on the received signal to be transmitted and the preset interference coefficient to obtain a plurality of first encoded signals; Using each of the power distribution networks respectively, encoding is performed based on the preset interference coefficient to obtain a plurality of second coded signals; Respectively using the coding modulation layer, based on the first coded signal obtained by the encoder network coding in the signal transmission module corresponding to each of the signal transmission modules, and the second coded signal obtained by the power allocation network coding, obtain a transmission signal transmitted by the signal transmission module corresponding to each of the signal transmission modules; Utilizing the channel layer, based on preset noise, the preset interference coefficient and each of the transmission signals, calculating an input signal of the signal output module corresponding to each of the signal transmission modules; The signal output modules corresponding to the signal transmission modules are respectively used to decode the input signals to obtain a plurality of output signals.

2. The communication method based on the network communication model according to claim 1, characterized in that: The first encoder network includes a first input layer, a first encoding layer, a first connection layer, a second encoding layer and a normalization layer, the first encoder network is any one of the plurality of encoder networks, and each of the encoder networks is used to encode based on the received signal to be transmitted and the preset interference coefficient to obtain a plurality of first encoded signals, including: Using the first input layer to align the received signal to be transmitted and the preset interference coefficient; The signal to be transmitted after the alignment processing is encoded by using the first coding layer to obtain a first intermediate coded signal, where the first intermediate coded signal includes multiple first activation values: Using the first connection layer to connect the first activation values, to obtain a second intermediate transmission signal; encoding the second intermediate transmission signal by using the second coding layer to obtain a third intermediate transmission signal; Using the normalization layer to perform normalization processing based on the third intermediate transmission signal to obtain the first coded signal; Based on the preset interference coefficient, after channel equalization is performed on the target signal output by the target input layer, the target signal is input to the next target input layer adjacent to the target input layer, and the target input layer includes any layer among the input layer, the first coding layer, the first connection layer, the second coding layer and the normalization layer.

3. The communication method based on the network communication model according to claim 2, characterized in that: The first input layer includes a plurality of input nodes, the first coding layer includes a plurality of activation functions, and the first coding layer is used to encode the signal to be transmitted after the alignment process to obtain a first intermediate coded signal, including: Calculating a first activation value based on the signals to be transmitted after the alignment processing transmitted by each input node using a first activation function, where the first activation function is any one of the plurality of activation functions; The activation function includes: Among them, φ(x j ) represents the first activation value, w b represents the first learning parameter, w s represents the second learning parameter, b(x j ) represents the basic activation function, B i (x j ) represents the spline function, c i represents the coefficient of the spline function, i represents the number of the spline function, x j Represents the signal to be transmitted after the alignment processing transmitted by the j-th input node.

4. The communication method based on the network communication model according to claim 1, characterized in that: The calculating the input signal of the signal output module corresponding to each of the signal transmission modules by using the channel layer based on the preset noise, the preset interference coefficient and each of the transmission signals includes: The channel layer is used to calculate the input signal of the signal output module corresponding to each signal transmission module based on the preset noise, the preset interference coefficient and each of the transmission signals by the following formula: in, represents the input signal of the output module corresponding to the mth signal transmission module, represents the transmission signal transmitted by the signal transmission module corresponding to the m-th signal transmission module, α represents the preset interference coefficient, represents the transmission signal transmitted by the signal transmission module corresponding to the nth signal transmission module that interferes with the mth signal transmission module, N represents the number of signal transmission modules that interfere with the mth signal transmission module, Represents the preset noise of the mth signal transmission module.

5. The communication method based on the network communication model according to claim 1, characterized in that: The first encoder network includes a first encoding layer, the first encoder network is any one of the plurality of encoder networks, and before encoding is performed based on the received signal to be transmitted and the preset interference coefficient using each encoder network respectively, the method further includes: Initializing the network communication model to be trained based on preset parameters, the parameters including: a first number of the transmission signal data sets to be trained, a second number of the transmission signals to be trained processed by the network communication model to be trained, a third number of channels, the number of times the network communication model is updated, the number of activation functions in the first coding layer, the order and grid precision of the spline function in the activation function, and a learning rate of the network communication model to be trained; Based on the third number, cyclically training the network communication model to be trained; If the network communication model reaches a convergence condition, the network communication model is obtained.

6. The communication method based on the network communication model according to claim 5, characterized in that: The training of the network communication model to be trained includes: Randomly generate interference coefficients to be trained and transmission signal data sets to be trained; Generate noise to be trained based on a predefined search space; Freeze channel gradients; Based on the interference coefficient to be trained, the transmission signal data set to be trained and the noise to be trained, the network communication model to be trained is trained to obtain an output signal data set to be trained; Calculating a loss value based on the transmission signal data set to be trained and the output signal data set to be trained; Based on the loss value, the weights of the network communication model to be trained are updated using back propagation technology.

7. The communication method based on the network communication model according to claim 5, characterized in that: The calculating the loss value based on the transmission signal data set to be trained and the output signal data set to be trained includes: Based on the transmission signal data set to be trained and the output signal data set to be trained, the loss value is calculated by the following loss function: in, Represents the loss value, L i represents the first training temporary value, G represents the number of transmission signal data sets to be trained or output signal data sets to be trained, represents the i-th transmission signal to be trained, Represents the i-th output signal to be trained.

8. A communication device based on a network communication model, characterized in that: The network communication model includes a channel layer and a plurality of signal transmission modules, each of which includes a signal transmission module and a signal output module; each of which includes an encoder network, a power distribution network and a coding modulation layer, and the communication device includes: A first encoding module, configured to respectively use each of the encoder networks to encode based on the received signal to be transmitted and a preset interference coefficient to obtain a plurality of first encoded signals; A second encoding module, configured to respectively use each of the power distribution networks to perform encoding based on the preset interference coefficient to obtain a plurality of second encoded signals; A coding and modulation module, configured to obtain, by using the coding and modulation layer respectively, the first coded signal obtained by the encoder network coding in the signal transmission module corresponding to each of the signal transmission modules, and the second coded signal obtained by the power allocation network coding, a transmission signal transmitted by the signal transmission module corresponding to each of the signal transmission modules; A calculation module, configured to calculate an input signal of the signal output module corresponding to each of the signal transmission modules based on a preset noise, the preset interference coefficient and each of the transmission signals by using the channel layer; The decoding module is used to respectively utilize the signal output modules corresponding to the signal transmission modules to decode the input signals to obtain multiple output signals.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images