Ota method for reducing the papr of ofdm encrypted signals based on an autoencoder and system

Abstract

The application provides an OFDM encrypted signal PAPR suppression method and system based on a self-encoder, which comprises the following steps: collecting a data sequence of an OFDM system, splitting a complex signal of the data sequence into a real part and an imaginary part to obtain an input sequence; inputting a chaotic neural network; allowing the chaotic neural network to perform chaotic encryption on the input sequence to obtain encrypted data; inputting the encrypted data into an encoder, allowing the encoder to extract features of the encrypted data and compress the features to a low latitude to obtain a low latitude result; inputting the low latitude result into a decoder to obtain a reconstruction result; performing inverse fast Fourier transform on the reconstruction result to obtain a time domain signal; converting the time domain signal into a frequency domain form for reconstruction to recover an original signal; and restoring the data sequence according to the original signal. The application has a significant improvement in reducing the performance of an encrypted OFDM signal, and the improvement is 10.7 dB compared with a non-suppression algorithm and 2.1 dB compared with an existing PRNet scheme.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication and signal processing technology, specifically relating to a method and system for PAPR suppression of OFDM encrypted signals based on an autoencoder. More specifically, this method combines the advantages of chaotic neural networks and autoencoders to effectively reduce the peak-to-average power ratio of OFDM signals, ensure the system's bit error rate performance, and improve the security performance of the OFDM encryption system. Background Technology

[0002] OFDM technology is a multi-carrier transmission technology widely used in wireless communication systems, which has advantages such as combating frequency-selective fading and effectively utilizing spectrum resources.

[0003] However, OFDM systems exhibit high PAPR (Programmable Response Rate), leading to low power amplifier efficiency and easily causing signal nonlinear distortion, thus severely impacting system transmission performance. To address this, numerous methods have been proposed to reduce the PAPR of OFDM systems, including traditional clipping methods, partial transmission sequence methods, and deep learning-based PAPR suppression methods.

[0004] Specifically, chaotic systems possess randomness, initial condition sensitivity, and nonlinearity, characteristics that give them immense potential in communication encryption and signal processing. Chaotic encryption, by leveraging the high sensitivity and complex dynamics of chaotic systems, can effectively mask signals in OFDM systems, enhancing system security. Meanwhile, chaotic neural networks combine the characteristics of chaotic systems with those of artificial neural networks, introducing chaotic dynamics into neural networks, resulting in stronger nonlinear information processing capabilities and contributing to better performance in encryption and feature extraction.

[0005] Existing deep learning-based PRNet schemes have achieved some success in PAPR suppression in OFDM systems, but they do not consider novel physical layer encryption systems, resulting in limited control over the signal bit error rate. Furthermore, they suffer from shortcomings in neural network architecture and loss function design, making it difficult to simultaneously balance PAPR suppression and OFDM system security. Therefore, this paper proposes a novel PAPR suppression method based on chaotic encryption and autoencoders to improve upon existing PRNet schemes, aiming to better balance PAPR suppression performance, system bit error rate performance, and communication security.

[0006] Patent document CN114915398A discloses a chaotic encryption method, application, computer device, and storage medium based on a variational autoencoder. The scheme includes acquiring an original chaotic sequence; inputting the original chaotic sequence data into a pre-trained variational autoencoder model for reconstruction processing, outputting a reconstructed chaotic sequence; and performing phase masking encryption and frequency masking encryption on communication data based on the obtained reconstructed chaotic sequence. The variational autoencoder model includes an encoder network and a decoder network. The encoder network includes alternating residual connection blocks and self-attention layers, and the decoder network includes convolutional layers and a loss layer placed after each convolutional layer. By introducing a variational autoencoder model, the generation rules of chaotic sequences are learned, replacing the inefficient generation of chaotic sequences through linear solution of differential equations. Through a large amount of chaotic sequence training data, the model's performance and generalization abilities are trained, effectively improving the generation efficiency of chaotic sequences. However, this scheme has limited control over the signal bit error rate and certain shortcomings in the architecture and loss function design of the neural network, making it difficult to simultaneously balance PAPR suppression and OFDM system security performance. This problem urgently needs to be solved. Summary of the Invention

[0007] To address the shortcomings of existing technologies, the purpose of this invention is to provide a method and system for suppressing the PAPR of OFDM encrypted signals based on an autoencoder.

[0008] A method for suppressing the PAPR of OFDM encrypted signals based on an autoencoder, according to the present invention, includes:

[0009] Step S1: Acquire the data sequence of the OFDM system, decompose the complex signal of the data sequence into real and imaginary parts, and obtain the input sequence based on the real and imaginary parts;

[0010] Step S2: Input the input sequence into a chaotic neural network; let the chaotic neural network perform chaotic encryption on the input sequence to obtain encrypted data;

[0011] Step S3: Input the encrypted data into the encoder, and have the encoder extract features from the encrypted data and compress it to a low dimension to obtain a low-dimensional result; input the low-dimensional result into the decoder to obtain the reconstruction result;

[0012] Step S4: Perform an inverse fast Fourier transform on the reconstructed result to obtain the time-domain signal;

[0013] Step S5: Convert the time-domain signal into frequency-domain form for reconstruction to restore the original signal; restore the data sequence in step S1 based on the original signal.

[0014] Preferably, in step S1, the real part and the imaginary part are arranged in the order of splitting;

[0015] In step S2, the initial values ​​of the chaotic neural network, namely x1, x2, x3, x4, x5, x6, are set as keys. The keys are input into the ode45 function, and the function is iteratively solved to calculate the sum of the first set of chaotic sequences and take the remainder of the sum with 6 to obtain the encrypted data.

[0016] Preferably, in step S3, the encrypted data is mapped using linear transformation and activation function to obtain a low-dimensional result;

[0017] The low-dimensional result is the encrypted data after linear transformation and activation function mapping to remove signal redundancy; the activation function includes: the Sigmoid function or the ReLU function;

[0018] In step S4, the time-domain signal is converted from digital to analog and then wirelessly transmitted.

[0019] Preferably, in step S3, the encoder and the decoder determine the loss function of the chaotic neural network; the mathematical expression of the loss function is:

[0020]

[0021] in, λ represents the loss function of the neural network, which is jointly determined by the encoder and decoder; λ represents a preset parameter; X represents the original input data. This represents the output of the decoder;

[0022]

[0023] Where b represents another preset parameter; This represents the impairment of the signal after passing through the channel, and a loss function is defined based on this difference value; L1, i.e. The mathematical expression is:

[0024]

[0025] Where X represents the signal point transmitted by the transmitter; The output signal point of the decoder is represented by FFF, which represents Fast Fourier Transform; H represents the received data after the signal has been affected by the channel; f represents the output function of the encoder, which represents Inverse Fast Fourier Transform; g represents the output function of the decoder; and ||||2 represents the L2 norm.

[0026] According to the present invention, an OFDM encryption signal PAPR suppression system based on an autoencoder includes:

[0027] Module M1: Acquires the data sequence of the OFDM system, decomposes the complex signal of the data sequence into real and imaginary parts, and obtains the input sequence based on the real and imaginary parts;

[0028] Module M2: Input the input sequence into a chaotic neural network; instruct the chaotic neural network to perform chaotic encryption on the input sequence to obtain encrypted data;

[0029] Module M3: Input the encrypted data into the encoder, instruct the encoder to extract features from the encrypted data and compress it to a low dimension to obtain a low-dimensional result; input the low-dimensional result into the decoder to obtain the reconstruction result;

[0030] Module M4: Performs an inverse fast Fourier transform on the reconstructed result to obtain the time-domain signal;

[0031] Module M5: Converts the time-domain signal into frequency-domain form for reconstruction to restore the original signal; restores the data sequence in module M1 based on the original signal.

[0032] Preferably, in module M1, the real part and the imaginary part are arranged in the order of splitting;

[0033] In module M2, the initial values ​​of the chaotic neural network, namely x1, x2, x3, x4, x5, x6, are set as keys. The keys are input into the ode45 function, and the function iteratively solves the problem to calculate the sum of the first set of chaotic sequences and take the remainder of the sum with 6 to obtain the encrypted data.

[0034] Preferably, in module M3, the encrypted data is mapped through linear transformation and activation function to obtain low-dimensional results;

[0035] The low-dimensional result is the encrypted data after linear transformation and activation function mapping to remove signal redundancy; the activation function includes: the Sigmoid function or the ReLU function;

[0036] In module M4, the time-domain signal is converted from digital to analog and then wirelessly transmitted.

[0037] Preferably, in module M3, the encoder and the decoder determine the loss function of the chaotic neural network; the mathematical expression of the loss function is:

[0038]

[0039] in, λ represents the loss function of the neural network, which is jointly determined by the encoder and decoder; λ represents a preset parameter; X represents the original input data. This represents the output of the decoder;

[0040]

[0041] Where b represents another preset parameter; This represents the impairment of the signal after passing through the channel, and a loss function is defined based on this difference value; L1, i.e. The mathematical expression is:

[0042]

[0043] Where X represents the signal point transmitted by the transmitter; denoted as the output signal point of the decoder; FFT represents Fast Fourier Transform; H represents the received data after the signal has passed through the channel; f represents the output function of the encoder; IFFT represents the Inverse Fast Fourier Transform; g represents the output function of the decoder; ||||2 represents the L2 norm.

[0044] According to the present invention, a computer-readable storage medium storing a computer program is provided, wherein when the computer program is executed by a processor, the steps of the autoencoder-based OFDM encrypted signal PAPR suppression method are implemented.

[0045] An electronic device according to the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of the autoencoder-based OFDM encrypted signal PAPR suppression method.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. This invention discloses an encrypted orthogonal frequency-division multiplexing (OFDM) signal peak-to-average power ratio (PAPR) suppression method based on an autoencoder. This invention utilizes chaotic encryption to effectively protect the signal and uses an autoencoder to extract and reconstruct features from the encrypted signal, thereby reducing the PAPR of the signal in the encrypted OFDM system.

[0048] 2. This invention does not require the design of additional complex accessories. It can achieve low PAPR and high security signal transmission through a simple encryption and self-encoding process.

[0049] 3. The loss function used in this invention includes multiple optimization terms that control bit error rate, PAPR and encryption security, ensuring that the system maintains a low bit error rate and high encryption strength while achieving PAPR suppression.

[0050] 4. This invention has a simple structure and high computational efficiency. It can significantly improve security performance and improve the transmission performance of the system in complex wireless communication environments. It has a significant improvement in reducing PAPR performance of encrypted OFDM signals. Compared with the method without suppression algorithm, it improves by 10.7dB and compared with the existing PRNet scheme by 2.1dB. Attached Figure Description

[0051] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0052] Figure 1 This is a flowchart provided for the present invention;

[0053] Figure 2 A schematic diagram of the phase space of the chaotic neural network provided by the present invention;

[0054] Figure 3 A schematic diagram of the encoder and decoder provided by the present invention;

[0055] Figure 4 This is a block diagram of the internal structure of the encoder provided by the present invention. Detailed Implementation

[0056] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0057] Since combining the loss function and network structure of chaotic neural networks and autoencoders leads to an unstable PAPR state in the encrypted system, this invention aims to address the problem that existing PAPR suppression methods for OFDM systems have a significant impact on the system's bit error rate and insufficient performance optimization. It proposes a PAPR suppression method based on chaotic encryption and autoencoders. This method intelligently learns the intrinsic characteristics of the input signal by combining chaotic neural networks and autoencoder networks in the OFDM system. This allows for encryption and PAPR suppression at the transmitting end using an encoder, and for recovering the original signal at the receiving end using a decoder.

[0058] The improvements in this invention include optimizations to the loss function and model structure of the chaotic neural network and autoencoder. Regarding the loss function design, this invention proposes a new form, comprising three parts: a term for controlling bit error rate performance, a term for controlling PAPR performance, and a term for encryption security. During training, once the system bit error rate reaches the expected target value, the loss function will further tend to reduce PAPR while ensuring the security of chaotic encryption. Furthermore, this invention improves the design of the autoencoder structure by replacing the activation function of the last layer of the autoencoder with eLU to improve signal reconstruction and ensure that the decoder can completely reconstruct the input data, thereby further optimizing PAPR suppression performance.

[0059] A method for suppressing the PAPR of OFDM encrypted signals based on an autoencoder, according to the present invention, includes:

[0060] S1: Signal preprocessing. This step modulates the raw data sequence of the OFDM system, splitting the complex signal into real and imaginary parts, and combining them sequentially to form the input sequence of the autoencoder. This step ensures that the input signal is suitable for subsequent autoencoder processing.

[0061] S2: Chaotic Neural Network Encryption. The input data sequence is fed into a chaotic neural network, which employs an activation function combining a multi-frequency cosine signal (MFCS) and a piecewise linear function. This network performs chaotic encryption on the input signal and outputs encrypted data.

[0062] Specifically, by calling the ode45 function in MATLAB to solve the differential equation, the chaotic system will continuously iterate through x1, x2, x... 3, The values ​​of x4, x5, and x6 are determined by setting the computational step size of the ode45 algorithm without altering its time approximation, thus selecting the number of iterations to solve the problem. This changes the length of the chaotic sequence, and the computational results for different number of iterations are observed. That is, 6 sets of non-independent chaotic sequences. Figure 2 It is the phase space diagram of the generated chaotic sequence, and the initial values ​​of the chaotic system, i.e., x1, x2, x... 3, x4, x5, and x6 are the keys. Calculate the sum of the first chaotic sequence and take the remainder when divided by 6. The resulting number is the key. XOR the signal to obtain the encrypted signal. Because this encrypted signal affects the distribution of the OFDM signal, the PAPR value will be affected to some extent.

[0063] S3: Encoder processing, inputting the chaotically encrypted data into the autoencoder model, such as... Figure 3As shown, an autoencoder consists of two parts: an encoder and a decoder. First, the encoder extracts features from the data, compressing the high-dimensional input into a low-dimensional representation. Since the original input is a high-dimensional, complex, and redundant signal, leading to high computational complexity, it is mapped to a low-dimensional vector through linear transformation and activation functions. This compressed representation is then input into the decoder for reconstruction. The loss function is defined as a representation related to the bit error rate. At the decoder, an improved eLU activation function is used to improve the accuracy of signal reconstruction. The activation function includes either the Sigmoid function or the ReLU function.

[0064] In an autoencoder, the loss function of the neural network is defined as

[0065]

[0066] in, λ represents the loss function of the neural network, which is jointly determined by the encoder and decoder; λ represents a preset parameter; X represents the signal point transmitted by the transmitter. Indicates the output signal point of the decoder;

[0067] Where L2(X) represents the PAPR loss value of the signal; The loss value before and after encoding is expressed as:

[0068]

[0069] Where b represents another preset parameter. This represents the impairment of the signal after passing through the channel, and a loss function is defined based on this difference value, where... The definition is as follows:

[0070]

[0071] In other words, That is, the mathematical expression for L1 is:

[0072]

[0073] Where X represents the signal point transmitted by the transmitter. The decoder output is represented by FFT, which stands for Fast Fourier Transform; H represents the received data after the signal has passed through the channel; f represents the encoder output function, which stands for Inverse Fast Fourier Transform; MSE represents the mean square error; g represents the decoder's generator function; ||||2 represents the L2 norm, which is the square root of the sum of squares of a given vector.

[0074] L2= APR(x(n))= APR(IfFT(f(X))) (4)

[0075] Where L2 represents the PAPR value of the signal after transmission; PAPR represents the peak-to-average power ratio of the input signal; x(n) represents the original input signal at n points, where n represents n signal points; IFFT represents the inverse fast Fourier transform.

[0076] Therefore, the loss of the neural network is related to the PAPR value, and the direction of gradient descent of the neural network is consistent with the direction of PAPR value descent.

[0077] In the formula, the parameters λ and b are manually set and can be changed according to the specific application. When L1 is greater than b, the system bit error rate does not achieve the expected effect. According to formula (2), L3 is not 0, and L is still related to the system bit error rate. When changing network parameters after training, the bit error rate performance needs to be considered. When L1 is less than b, it means that the system bit error rate has been optimized to below b. According to formula (2), L3 is equal to 0. At this point, it only relates to the system λL2(X) value, while L2 is related to PAPR((())), that is, it is related to PAPR. During the training phase, it will tend to reduce PAPR to maximize its PAPR suppression effect.

[0078] Specifically, an OFDM system has N subcarriers, and the signal on the i-th subcarrier is x'. i Since the signal is complex, the sequence to be input to the encoder is X' = [x'1, x'2, ..., x']. N ].like Figure 4 As shown, since the autoencoder does not accept complex inputs, the input sequence needs to be processed before being input to the encoder, by splitting each signal in X' into its real part x'. Re,i With the imaginary part x' Im,i And arrange them in a one-to-one correspondence order. Similarly, to recombine the real and imaginary parts into a complex signal, the sequence y' is required. i Let f(X) be the frequency domain signal on the i-th subcarrier. f(X) is then transmitted through the N-point IFFT module and DAC module. At the receiver, the received signal is processed by the ADC module and FFT module, and the signal input to the decoder is FFT(H*IFFT(f(X))).

[0079] The trained decoder will reconstruct the input sequence based on the input. Conversely to the operation on the receiver, it is necessary to... The sequence is recombined into complex numbers in order, and then the output sequence is obtained through parallel-to-serial conversion. The original data can then be obtained through demodulation.

[0080] S4: Inverse Fast Fourier Transform (IFFT). In S3, the input signal is encoded using a trained encoder to obtain a low PAPR signal. This signal is then subjected to an Inverse Fast Fourier Transform (IFFT) to convert it into a time-domain signal. Subsequently, it is transmitted through a digital-to-analog converter (DAC) to the channel. The purpose of this step is to convert the signal into a form suitable for wireless transmission.

[0081] S5: Receiver signal processing. At the receiver, the received time-domain signal is first processed by analog-to-digital conversion (ADC), followed by Fast Fourier Transform (FFT) to convert the signal back to frequency domain form. This step ensures that the signal can be correctly processed by the decoder.

[0082] S6: Decoding and decryption. The frequency domain signal processed by FFT is input into the decoder obtained in S3 to reconstruct the signal and recover the original signal. Then, the original data signal is recovered by decryption through the inverse process of a chaotic neural network.

[0083] S7: Signal restoration. The decoded signal is recombined into a complex form, demodulated, and the final original data sequence is obtained. Through this process, the original data from the transmitting end is successfully recovered, ensuring the communication quality of the system.

[0084] Specifically, in step S1, the real and imaginary parts of the data after splitting are arranged in order to ensure that the components of each complex signal are correctly input into the autoencoder and encrypted one by one according to the subsequent chaotic sequence.

[0085] Specifically, in step S2, the activation function of the chaotic neural network is composed of multiple single-frequency cosine signals superimposed to form a complex nonlinear mapping relationship, thereby ensuring the chaos and security of the encryption.

[0086] Specifically, in step S3, the activation function of the last layer of the autoencoder is eLU to ensure that the negative part of the signal can be correctly reconstructed, thereby enhancing the accuracy of signal restoration. The loss function is also reconstructed to ensure that the PAPR value can be suppressed to the maximum extent.

[0087] Specifically, in step S4, the frequency domain signal is converted into a time domain signal using the standard IFFT algorithm, ensuring that the signal can adapt to the transmission requirements of the OFDM system after being processed by chaotic encryption and autoencoder.

[0088] Specifically, in step S6, the chaotic neural network decryption process uses parameter settings that are the opposite of those in the encryption process, ensuring that the legitimate receiver can correctly restore the original data, while the illegitimate receiver cannot decipher the signal.

[0089] Furthermore, the method of this invention employs a dynamic learning rate and the Adam optimizer for training, ensuring that the model can converge quickly and effectively during training, thereby achieving optimal suppression of PAPR and maintaining signal security without degrading the system bit error rate.

[0090] The present invention also provides an autoencoder-based OFDM encryption signal PAPR suppression system. The autoencoder-based OFDM encryption signal PAPR suppression system can be implemented by executing the process steps of the autoencoder-based OFDM encryption signal PAPR suppression method. That is, those skilled in the art can understand the autoencoder-based OFDM encryption signal PAPR suppression method as a preferred embodiment of the autoencoder-based OFDM encryption signal PAPR suppression system.

[0091] According to the present invention, an OFDM encryption signal PAPR suppression system based on an autoencoder includes:

[0092] Module M1: Acquires the data sequence of the OFDM system, decomposes the complex signal of the data sequence into real and imaginary parts, and obtains the input sequence based on the real and imaginary parts;

[0093] Module M2: Input the input sequence into a chaotic neural network; instruct the chaotic neural network to perform chaotic encryption on the input sequence to obtain encrypted data;

[0094] Module M3: Input the encrypted data into the encoder, instruct the encoder to extract features from the encrypted data and compress it to a low dimension to obtain a low-dimensional result; input the low-dimensional result into the decoder to obtain the reconstruction result;

[0095] Module M4: Performs an inverse fast Fourier transform on the reconstructed result to obtain the time-domain signal;

[0096] Module M5: Converts the time-domain signal into frequency-domain form for reconstruction to restore the original signal; restores the data sequence in module M1 based on the original signal.

[0097] Those skilled in the art will understand that, besides implementing the system and its various devices, modules, and units provided by this invention in the form of purely computer-readable program code, the same functions can be achieved entirely through logical programming of the method steps, making the system and its various devices, modules, and units of this invention function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, the system and its various devices, modules, and units provided by this invention can be considered as a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; alternatively, the devices, modules, and units for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0098] In the description of this application, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0099] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A method for suppressing PAPR (Parasyte Propagation Reduction) of OFDM encrypted signals based on an autoencoder, characterized in that, include: Step S1: Acquire the data sequence of the OFDM system, decompose the complex signal of the data sequence into real and imaginary parts, and obtain the input sequence based on the real and imaginary parts; Step S2: Input the input sequence into the chaotic neural network; The chaotic neural network is used to perform chaotic encryption on the input sequence to obtain encrypted data; Step S3: Input the encrypted data into the encoder, and have the encoder extract features from the encrypted data and compress it to a low dimension to obtain a low-dimensional result; input the low-dimensional result into the decoder to obtain the reconstruction result; Step S4: Perform an inverse fast Fourier transform on the reconstructed result to obtain the time-domain signal; Step S5: Convert the time-domain signal into frequency-domain form for reconstruction to restore the original signal; restore the data sequence from step S1 based on the original signal; In step S1, the real part and the imaginary part are arranged in the order of splitting; In step S2, the initial values ​​of the chaotic neural network, namely x1, x2, x3, x4, x5, x6, are set as keys. The keys are input into the ode45 function, and the function is iteratively solved to calculate the sum of the first set of chaotic sequences and take the remainder of the sum with 6 to obtain the encrypted data. In step S3, the encoder and the decoder determine the loss function of the chaotic neural network; the mathematical expression of the loss function is: （1） in, This represents the loss function of the neural network, which is determined jointly by the encoder and decoder. This represents a preset parameter; This represents the original input data; This represents the output of the decoder; This represents the PAPR loss value of the signal; This represents the loss value before and after encoding; the stated The mathematical expression is: （2） in, This indicates another preset parameter; This represents the impairment of the signal after passing through the channel, and a loss function is defined based on this difference value; The mathematical expression is: in, Indicates the signal point transmitted by the transmitter. This represents the output of the decoder. Indicates Fast Fourier Transform; This represents the received data after the signal has been affected by the channel. This represents the encoder's output function. Indicates the inverse fast Fourier transform; represents the generator function of the decoder; ||||2 represents the L2 norm.

2. The method for suppressing the PAPR of OFDM encrypted signals based on an autoencoder according to claim 1, characterized in that, In step S3, the encrypted data is mapped using linear transformation and activation function to obtain a low-dimensional result; The low-dimensional result is the encrypted data after linear transformation and activation function mapping to remove signal redundancy; The activation function includes: the Sigmoid function or the ReLU function; In step S4, the time-domain signal is converted from digital to analog and then wirelessly transmitted.

3. A PAPR suppression system for OFDM encrypted signals based on an autoencoder, characterized in that, include: Module M1: Acquires the data sequence of the OFDM system, decomposes the complex signal of the data sequence into real and imaginary parts, and obtains the input sequence based on the real and imaginary parts; Module M2: Inputs the input sequence into the chaotic neural network; The chaotic neural network is used to perform chaotic encryption on the input sequence to obtain encrypted data; Module M3: Input the encrypted data into the encoder, which extracts features from the encrypted data and compresses it to a low dimension to obtain a low-dimensional result; input the low-dimensional result into the decoder to obtain the reconstruction result; Module M4: Performs an inverse fast Fourier transform on the reconstructed result to obtain the time-domain signal; Module M5: Converts the time-domain signal into frequency-domain form for reconstruction to restore the original signal; restores the data sequence in module M1 based on the original signal; In module M1, the real part and the imaginary part are arranged in the order of splitting; In module M2, the initial values ​​of the chaotic neural network, namely x1, x2, x3, x4, x5, and x6, are set as keys. The keys are input into the ode45 function, and the function is iteratively solved to calculate the sum of the first set of chaotic sequences and take the remainder of the sum with 6 to obtain the encrypted data. In module M3, the encoder and the decoder determine the loss function of the chaotic neural network; the mathematical expression of the loss function is: （1） in, This represents the loss function of the neural network, which is determined jointly by the encoder and decoder. This represents a preset parameter; This represents the original input data; This represents the output of the decoder; （2） in, This indicates another preset parameter; This represents the impairment of the signal after passing through the channel, and a loss function is defined based on this difference value; The mathematical expression is: in, Indicates the signal point transmitted by the transmitter. This represents the output of the decoder. Indicates Fast Fourier Transform; This represents the received data after the signal has been affected by the channel. This represents the encoder's output function. Indicates the inverse fast Fourier transform; represents the generator function of the decoder; ||||2 represents the L2 norm.

4. The OFDM encryption signal PAPR suppression system based on an autoencoder according to claim 3, characterized in that, In module M3, the encrypted data is mapped through linear transformation and activation function to obtain a low-dimensional result; The low-dimensional result is the encrypted data after linear transformation and activation function mapping to remove signal redundancy; The activation function includes: the Sigmoid function or the ReLU function; In module M4, the time-domain signal is converted from digital to analog and then wirelessly transmitted.

5. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the autoencoder-based OFDM encrypted signal PAPR suppression method as described in any one of claims 1 to 2.

6. 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 computer program is executed by the processor, it implements the steps of the autoencoder-based OFDM encrypted signal PAPR suppression method as described in any one of claims 1 to 2.

Images