An OFDM chaotic communication method, apparatus, and medium with reference signal multiplexing.

Abstract

This invention provides an OFDM chaotic communication method, apparatus, device, and medium with reference signal multiplexing. The bit stream to be transmitted is divided into b... k Reference bit and b i Index bits, each reference bit triggers a generator group consisting of two heterogeneous chaotic mappings based on its 0 or 1 value, generating b. k A set of chaotic sequences of length β constitutes a reference sequence set R; b i The index bits are divided into G groups. Each group selects one chaotic sequence from R using an index selector as the data carrier sequence. The R and G data carrier sequences are then superimposed using OFDM modulation to form the transmitted signal. At the receiving end, the signal undergoes cyclic prefix removal, serial-to-parallel conversion, and FFT to separate the reference and index data portions. Finally, a deep neural network demodulation model recovers the data. k Reference bits and reconstruct R, then use the reconstructed R to perform cross-correlation detection on the index data to recover b. i Index bits. This allows the reference signal to synchronously complete reference transmission and user information transmission on the same time, frequency, and power resources, eliminating the pure overhead of the reference signal.

Description

Technical Field

[0001] This invention relates to the field of chaotic digital modulation and demodulation technology in wireless communication, and particularly to an OFDM chaotic communication method, apparatus, and medium with reference signal multiplexing. Background Technology

[0002] Chaotic signals possess noise-like characteristics, extreme sensitivity to initial conditions, and excellent autocorrelation and cross-correlation properties. Theoretically, they can generate an infinite number of uncorrelated sequences and are widely used in physical layer security and interference-resistant transmission scenarios in wireless communication. In existing incoherent chaotic communication schemes, due to the elimination of complex chaotic synchronization between the transmitting and receiving ends, the receiver structure is simple and the hardware implementation cost is low, making it the mainstream technology for civilian chaotic communication.

[0003] Existing incoherent chaotic communication schemes generally adopt a "reference-information" time-division (or carrier-division) frame structure: within each transmission symbol, a chaotic sequence without any user information is first transmitted as a reference signal using a time slot (or a set of subcarriers), and then a data signal modulated by the reference sequence is transmitted using another time slot (or another set of subcarriers); the receiving end uses the received reference signal and data signal to perform cross-correlation operations to complete demodulation. Under this frame structure, the reference signal and data signal are independent of each other in terms of time and frequency resources and do not reuse each other; the reference signal itself does not carry any user bits to be transmitted.

[0004] The direct consequence of the aforementioned frame structure is that the reference signal occupies approximately half of the symbol energy and time-frequency resources, yet contributes nothing to the transmission rate of user information. This theoretically limits the system's spectral efficiency and energy efficiency to less than half of the theoretical upper limit. Even with the introduction of OFDM to parallelize serial data, as long as the reference subcarrier and data subcarrier remain separate, this ratio cannot be substantially improved.

[0005] In view of the above, this application is hereby submitted. Summary of the Invention

[0006] This invention discloses an OFDM chaotic communication method, device, and medium with reference signal multiplexing, aiming to solve the problems in existing incoherent chaotic communication schemes where reference signals and data signals are transmitted separately in terms of time and frequency resources, the reference signal itself does not carry any user bits and thus occupies about half of the symbol energy and time and frequency resources, resulting in the system's spectral efficiency and energy efficiency being limited to less than half of the theoretical upper limit in principle.

[0007] The first embodiment of the present invention provides an OFDM chaotic communication method with reference signal multiplexing, comprising: At the transmitting end, the bit stream to be transmitted is divided into... one reference bit and One index bit; the... Each reference bit is sequentially input into a chaos generator group consisting of two heterogeneous chaotic maps. Each reference bit triggers the generation of the corresponding chaotic map based on its value. These chaotic sequences form a reference sequence set R; The The index bits are evenly divided into G groups, each group Each set of index bits selects one chaotic sequence from the reference sequence set R through an index selector as the data carrier sequence for that set, resulting in G data carrier sequences. The reference sequence set R and the G data carrier sequences are then superimposed after OFDM modulation to form a transmitted chaotic signal T, which is then sent into the wireless channel. The receiving end processes the received signal through cyclic prefix removal, serial-to-parallel conversion, and Fast Fourier Transform. Based on the subcarrier allocation scheme of the transmitting end, it separates the reference portion and the index data portion. The reference portion is then input into a pre-trained deep neural network demodulation model, which outputs... Reference bits, and the A reference sequence set R is reconstructed from the reference bits according to the mapping relationship; each data-carrying sequence in the index data part is cross-correlated with the reconstructed reference sequence set R, and the sequence position corresponding to the maximum correlation value is used to recover the sequence. Each index bit, thus restoring the entire... One index bit; will one reference bit and The index bits are merged in the order of partitioning to obtain the original transmission bit stream.

[0008] A second embodiment of the present invention provides an OFDM chaotic communication device with reference signal multiplexing, comprising: The reference sequence generation unit is used at the transmitting end to divide the bit stream to be transmitted into... one reference bit and One index bit; the... Each reference bit is sequentially input into a chaos generator group consisting of two heterogeneous chaotic maps. Each reference bit triggers the generation of the corresponding chaotic map based on its value. These chaotic sequences form a reference sequence set R; Indexed modulation and OFDM transmission unit, used to transmit the... The index bits are evenly divided into G groups, each group Each set of index bits selects one chaotic sequence from the reference sequence set R through an index selector as the data carrier sequence for that set, resulting in G data carrier sequences. The reference sequence set R and the G data carrier sequences are then superimposed after OFDM modulation to form a transmitted chaotic signal T, which is then sent into the wireless channel. The receiving demodulation unit is used to separate the reference part and the index data part of the received signal after performing cyclic prefix removal, serial-to-parallel conversion, and fast Fourier transform at the receiving end, according to the subcarrier allocation scheme at the transmitting end; the reference part is input into a pre-trained deep neural network demodulation model, and the output is... Reference bits, and the A reference sequence set R is reconstructed from the reference bits according to the mapping relationship; each data-carrying sequence in the index data part is cross-correlated with the reconstructed reference sequence set R, and the sequence position corresponding to the maximum correlation value is used to recover the sequence. Each index bit, thus restoring the entire... One index bit; will one reference bit and The index bits are merged in the order of partitioning to obtain the original transmission bit stream.

[0009] The third embodiment of the present invention provides an OFDM chaotic communication device with reference signal multiplexing, including a memory and a processor. The memory stores a computer program, which can be executed by the processor to implement an OFDM chaotic communication method with reference signal multiplexing as described in any of the above embodiments.

[0010] The fourth embodiment of the present invention provides a computer-readable storage medium, characterized in that it stores a computer program, which can be executed by a processor of the device in which the computer-readable storage medium is located, to implement an OFDM chaotic communication method for reference signal multiplexing as described in any of the above embodiments.

[0011] Based on the OFDM chaotic communication method, apparatus, and medium for reference signal multiplexing provided by this invention, the bit stream to be transmitted is divided into... one reference bit and Each reference bit, with its value of 0 or 1, triggers the corresponding chaotic mapping in a group of chaotic generators composed of two heterogeneous chaotic mappings, generating... A set of chaotic sequences of length β forms a reference sequence set R, which carries both the... The information of each reference bit is used as the codebook for subsequent index modulation; then... Each index bit is uniformly divided into G groups. Each group selects one chaotic sequence from the reference sequence set R as the data carrier sequence using an index selector. The reference sequence set R and the G data carrier sequences are then superimposed after OFDM modulation to form the transmitted signal. At the receiving end, after cyclic prefix removal, serial-to-parallel conversion, and fast Fourier transform, the reference part and the index data part are separated, and the signal is recovered by a deep neural network demodulation model. The reference sequence set R is reconstructed using reference bits, and then the index data is cross-correlation detected and recovered using the reconstructed R. Each index bit enables the reference signal to synchronously complete reference transmission and user information transmission on the same time-frequency and power resources, eliminating the pure overhead of the reference signal. Attached Figure Description

[0012] Figure 1 This is a flowchart illustrating an OFDM chaotic communication method for reference signal multiplexing provided in the first embodiment of the present invention; Figure 2 This is a structural diagram of the system transmitter provided by the present invention; Figure 3 This is a structural diagram of the system receiver provided by the present invention; Figure 4 This is an example diagram of the system transmission signal format provided by the present invention; Figure 5 This is a schematic diagram of the deep neural network demodulator architecture of the present invention; Figure 6 This is a simulation diagram provided by the present invention; Figure 7 This is a schematic diagram of a reference signal multiplexing OFDM chaotic communication device provided in the second embodiment of the present invention. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] To better understand the technical solution of the present invention, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0015] This invention discloses an OFDM chaotic communication method, device, and medium with reference signal multiplexing, aiming to solve the problems in existing incoherent chaotic communication schemes where reference signals and data signals are transmitted separately in terms of time and frequency resources, the reference signal itself does not carry any user bits and thus occupies about half of the symbol energy and time and frequency resources, resulting in the system's spectral efficiency and energy efficiency being limited to less than half of the theoretical upper limit in principle.

[0016] Please combine Figure 1 and Figure 2 The first embodiment of the present invention provides an OFDM chaotic communication method with reference signal multiplexing, comprising: S101, at the transmitting end, the bit stream to be transmitted is divided into... one reference bit and One index bit; the... Each reference bit is sequentially input into a chaos generator group consisting of two heterogeneous chaotic maps. Each reference bit triggers the generation of the corresponding chaotic map based on its value. These chaotic sequences form a reference sequence set R; In this embodiment, the transmitting end first divides the bit stream to be transmitted into two parts: reference bits and index bits according to a preset ratio, wherein the number of reference bits is denoted as . The number of index bits is denoted as The reference bit is used to trigger the chaos generator to generate a reference chaotic sequence, and the index bit is used to select the data carrier sequence from the reference chaotic sequence. The chaos generator group, composed of two heterogeneous chaotic mappings, includes chaos generator A and chaos generator B. Chaos generator A and chaos generator B generate a length of [length missing] using different mapping methods. The two mapping methods are independent of each other in terms of iteration rules and parameter settings, and the two types of chaotic sequences generated thereby have their own independent chaotic characteristics and good mutual uncorrelation. It should be noted that in other embodiments, any two mapping methods that meet the above independence requirements can be selected from the existing chaotic mappings according to the needs of the actual communication scenario, respectively as the mapping rules of chaotic generator A and chaotic generator B. There is no specific limitation on this, but these schemes are all within the protection scope of this invention.

[0017] Regarding the The reference bits are processed sequentially according to their bit order: The first reference bit is read; when this bit is 0, chaos generator A is triggered to generate a line of length... A chaotic sequence, when the value of this bit is 1, triggers the chaos generator B to generate a sequence of length . The chaotic sequence is thus obtained, and the first chaotic sequence is obtained. Then, the second reference bit is read and the second chaotic sequence is obtained according to the same rule. And so on until the first one is completed. The processing of each reference bit yielded a total of The length of each strip is The chaotic sequence is arranged in the order of the reference bits to form a reference sequence set. In this process, since chaos generator A corresponds to bit 0 and chaos generator B corresponds to bit 1, whether each chaotic sequence in the reference sequence set R is generated by chaos generator A or chaos generator B corresponds to whether the value of the reference bit at that position is 0 or 1. That is, the reference sequence set R implicitly contains... All the original information of each reference bit; at the same time, the resulting reference sequence set R will also serve as a candidate codebook for subsequent index selector index modulation for the index bit to select the data bearer sequence. Thus, the bearing of reference bits and the generation of reference signals are realized simultaneously on a single time-frequency and power resource, laying the foundation for eliminating the resource occupation of pure overhead reference signals in traditional schemes.

[0018] S102, the The index bits are evenly divided into G groups, each group Each set of index bits selects one chaotic sequence from the reference sequence set R through an index selector as the data carrier sequence for that set, resulting in G data carrier sequences. The reference sequence set R and the G data carrier sequences are then superimposed after OFDM modulation to form a transmitted chaotic signal T, which is then sent into the wireless channel. Please combine Figure 3 and Figure 4 After completing the reference sequence set After its generation, the transmitter... Each index bit is grouped and indexed modulated. Specifically, the index bits are... The index bits are evenly divided into G groups, each group containing 1 bit, the number of groups G satisfies G= By using serial-to-parallel conversion, it is ensured that each group of index bits can be uniquely mapped to a reference chaotic sequence. For each group of G groups of index bits, an index selector selects one chaotic sequence from the reference sequence set R as the data carrier sequence for that group. The specific process is as follows: first, select the group... Each index bit is converted from binary to its corresponding decimal value, and then a string of length is generated based on the decimal value. A binary selection vector is generated, wherein the binary selection vector has a value of 1 only at the position corresponding to the decimal value, and all other positions have a value of 0. Finally, the binary selection vector is multiplied by the reference sequence set R. Since the chaotic sequences in the reference sequence set R are arranged in order, the chaotic sequence corresponding to the position of 1 in the binary selection vector is the one that is selected. The selected index bits are used as the data bearer sequence corresponding to that group of index bits. The process of converting binary to decimal, generating a selection vector, and multiplying by R is performed sequentially on the G groups of index bits, resulting in a total of G data bearer sequences. Subsequently, the reference sequence set R and the resulting G data bearer sequences are allocated to various subcarriers preset by the OFDM system. After OFDM modulation, they are superimposed to form the final transmitted chaotic signal T, which is then transmitted through the wireless channel via the transmit antenna.

[0019] In practical wireless communication scenarios, the transmitted chaotic signal T, after being emitted by the antenna, will undergo multipath reflection and refraction due to environmental factors such as buildings and the ground, while also being superimposed with environmental noise. To verify the system's transmission performance in a realistic communication scenario, this embodiment uses a multipath Rayleigh fading channel model to model the system's transmission process. The multipath Rayleigh fading channel contains L independent propagation paths, where L is the maximum number of paths in the channel. The channel response of each path is expressed in complex form as follows: , where l is the path number and l = 1, 2, …, L, and j is the imaginary unit. This represents the amplitude fading coefficient of the l-th path, which follows a Rayleigh distribution. Let represent the random phase shift of the l-th path and follow a uniform distribution on [0, 2π). After the transmitted chaotic signal T is transmitted through the above L independent paths, the arrival times of different paths at the receiver differ, forming a multipath delay superposition signal at the receiver. Additive white Gaussian noise is further superimposed on this signal. Therefore, the mathematical model of the combined chaotic received signal r(k) obtained at the receiver is: Where k is the discrete-time sampling point number, and T(k) is the value of the transmitted combined chaotic signal at the k-th sampling point. Let be the propagation delay of the l-th path relative to the direct path. This indicates that the transmitted chaotic signal T is delayed via the l-th path. The value at the kth sampling point after sampling points. The additive white Gaussian noise at the k-th sampling point has a mean of 0 and a variance of . , The signal r(k) represents the power spectral density of the single-sided noise; the received signal r(k) is the input signal that is subsequently processed by the receiver, including cyclic prefix removal, serial-to-parallel conversion, and fast Fourier transform.

[0020] S103, after the receiving end performs cyclic prefix removal, serial-to-parallel conversion, and fast Fourier transform on the received signal, it separates the reference part and the index data part according to the subcarrier allocation scheme of the transmitting end; the reference part is input into a pre-trained deep neural network demodulation model, and the output is... Reference bits, and the A reference sequence set R is reconstructed from the reference bits according to the mapping relationship; each data-carrying sequence in the index data part is cross-correlated with the reconstructed reference sequence set R, and the sequence position corresponding to the maximum correlation value is used to recover the sequence. Each index bit, thus restoring the entire... One index bit; will one reference bit and The index bits are merged in the order of partitioning to obtain the original transmission bit stream.

[0021] Please combine Figure 5 In this embodiment, after the receiver obtains the received signal r(k) transmitted through a multipath Rayleigh fading channel, it first performs cyclic prefix removal and serial-to-parallel conversion on the received signal sequentially. Then, it converts the signal from the time domain to the frequency domain using a fast Fourier transform. Subsequently, according to the subcarrier allocation scheme preset by the transmitter, it separates the reference part carrying the reference chaotic sequence and the index data part carrying the data sequence from the frequency domain signal, and sends them to the deep neural network demodulation path and the cross-correlation detection path for processing, respectively. For the reference part, it is input into a pre-trained deep neural network demodulation model, which performs feature extraction and classification, and demodulates the output. The deep neural network demodulation model consists of a reference bit; it comprises an input layer, four one-dimensional convolutional layers, a self-attention layer, two bidirectional LSTM layers, two fully connected layers with random deactivation, and a Softmax output layer. In the specific demodulation process, the chaotic sequence of the reference part is first Z-score standardized in the input preprocessing stage to achieve a mean of 0 and a variance of 1. Then, the standardized sequence is sequentially fed into the four one-dimensional convolutional neural network. After each convolution operation, a ReLU activation function is applied and layer normalization is performed. The number of output channels and the kernel width of each layer increase progressively to extract local features of the reference chaotic sequence and output a local feature matrix H. Next, the local feature matrix H is fed into the self-attention layer, where attention weights are calculated using a learnable query vector Q to obtain a weighted feature matrix. This highlights the feature components that contribute more to the decision of the reference bit; subsequently, the weighted feature matrix is... The data is fed into a bidirectional LSTM layer, which contains two layers of bidirectional LSTM units with 128 hidden units each, to extract global temporal features from both the forward and reverse directions of the weighted features. Finally, the global temporal features output by the bidirectional LSTM layer are concatenated with the local feature matrix H output by the aforementioned four-layer convolutional neural network, and then fed into two fully connected layers with random deactivation to reduce the risk of overfitting. Finally, a Softmax output layer outputs the probability vector of the nth reference bit. ,in and Let represent the probabilities of the nth reference bit being 0 and 1, respectively. Then, perform a maximum probability decision on the probability vector, i.e., select... and The binary symbol corresponding to the one with the higher probability value is used as the decision result for that reference bit, and this is applied sequentially to all bits. Each reference bit is used for judgment, and the final demodulation yields all the results. One reference bit.

[0022] Before being put into use, the deep neural network demodulation model needs to undergo offline training to obtain reliable demodulation performance. The offline training process is as follows: First, training sample pairs are constructed. For each training reference bit, a chaotic sequence of length β is generated using either chaos generator A or chaos generator B described in step 1, depending on whether the bit value is 0 or 1. This chaotic sequence is then transmitted through a multipath Rayleigh fading channel simulation environment within a set signal-to-noise ratio range to obtain the received subsequence. ,Will With the corresponding binary tag Form a training sample pair ,in Repeat the above process until M training sample pairs are accumulated; then set the hyperparameters required for training, including the total number of training samples M, the training signal-to-noise ratio range, batch size, learning rate, and validation set ratio, and divide the M training sample pairs into a training set and a validation set according to the validation set ratio; during training, binary cross-entropy is used as the loss function to evaluate the difference between the predicted output of the deep neural network demodulation model and the true label, and the loss function is defined as: Where M is the number of training samples. Let i be the true label of the i-th training sample. The deep neural network demodulation model estimates the output of the i-th training sample. In the iterative optimization stage, an adaptive moment estimator optimizer is used to perform multiple rounds of iterative optimization of the model parameters in a three-stage loop: forward inference to calculate the predicted distribution, backpropagation to calculate the gradient based on the loss function, and parameter update, until the loss function converges. After each training cycle, the generalization loss of the model on the validation set is used as a monitoring indicator. When the validation loss does not decrease significantly within a set number of consecutive training cycles, the model is considered to have been sufficiently trained. The iteration is terminated and the model with the best validation performance is saved as the final deep neural network demodulation model for actual demodulation.

[0023] Demodulated using the aforementioned deep neural network demodulation model After obtaining the reference bits, the receiving end re-calls the corresponding chaos generator according to the mapping relationship described in step 1 (i.e., bit 0 corresponds to chaos generator A, bit 1 corresponds to chaos generator B), and reconstructs the reference sequence set R consistent with that of the transmitting end locally. Based on this, cross-correlation detection is performed on the index data part to recover the index bits. Specifically, the reconstructed reference sequence set R is constructed as a reference matrix. Each row of the reference matrix B is a chaotic sequence in the corresponding order of the reference sequence set R; for each data-carrying sequence in the index data part, it is constructed as a data matrix. , and according to The form is cross-correlated with the reference matrix B to obtain a length of [missing information]. The correlation value vector Z; find the position corresponding to the maximum absolute value among the components of the correlation value vector Z. Since the transmitting end index selector converts the index bits into decimal values ​​and then selects the chaotic sequence corresponding to the position from the reference sequence set R when generating the data bearer sequence, after the receiving end obtains the position corresponding to the maximum correlation value, it restores the position to its original value according to the inverse mapping of the binary to decimal conversion. Each index bit represents the demodulation result of that group of index bits. The process of constructing the data matrix, performing cross-correlation with the reference matrix B, finding the position of the maximum correlation value, and inverse mapping to recover the index bits is sequentially performed on the G groups of data-bearing sequences. By concatenating these groups in order, the entire sequence can be recovered. Each index bit. Finally, the deep neural network demodulation model demodulates the resulting... The reference bits and cross-correlation detection recovered the By merging the index bits according to the division order at the transmitting end, the complete original transmitted bit stream can be obtained, thus completing the entire signal demodulation and information recovery process at the receiving end.

[0024] To verify the effectiveness of the OFDM chaotic communication method with reference signal multiplexing proposed in this embodiment, the inventors conducted a bit error rate (BER) performance simulation experiment under multipath Rayleigh fading channel conditions and compared it with existing OFDM-DCSK and DL-OFDM-DCSK schemes. The simulation parameters were set as follows: chaotic sequence length β=64, number of independent propagation paths L=3 in the multipath Rayleigh fading channel, and time delays of the three paths as follows. , , There are 2 sampling points, and the average power gain of each path is equal. The index bits are 2, divided into 2 groups, the reference bits are 4, and the total number of subcarriers is 6. Please combine this with... Figure 6 The horizontal axis in the figure represents the signal-to-noise ratio per bit, and the vertical axis represents the bit error rate (BER), using a logarithmic coordinate system. Figure 6 The figure shows a comparison of the bit error rate (BER) performance of this embodiment (labeled RSR-OFDM-DCSK) with the traditional OFDM-DCSK scheme and the DL-OFDM-DCSK scheme in a simulation environment. As can be seen from the figure, the traditional OFDM-DCSK scheme, due to the separate transmission of the reference signal and data signal and the receiver's reliance on cross-correlation detection for demodulation, exhibits the worst BER performance in multipath Rayleigh fading channels. Even when the signal-to-noise ratio per bit reaches 30 dB, its BER remains below 10. -2Near the order of magnitude, convergence was not achieved effectively. The DL-OFDM-DCSK scheme introduces deep learning-assisted demodulation, which significantly improves the BER performance compared to the traditional OFDM-DCSK scheme. The simulation results show that the deep multiplexing mechanism of the reference signal eliminates the pure overhead of the reference signal in the traditional scheme, enabling the transmission of more user bits under the same subcarrier resources and improving spectral efficiency. At the same time, the asymmetric hybrid demodulation architecture combining deep neural network and cross-correlation detection adopted by the receiver can effectively guarantee high-precision demodulation of the reference bit under multipath Rayleigh fading channel conditions, thereby providing an accurate reference matrix for cross-correlation detection of the index bit. This avoids the error cascade effect of reference bit misjudgment to index bit decision, making the overall bit error rate of the system better than the existing scheme across the entire signal-to-noise ratio range, verifying the effectiveness of this technical scheme in complex channel environments.

[0025] Please combine Figure 7 The second embodiment of the present invention provides an OFDM chaotic communication device with reference signal multiplexing, comprising: Reference sequence generation unit 201 is used at the transmitting end to divide the bit stream to be transmitted into... one reference bit and One index bit; the... Each reference bit is sequentially input into a chaos generator group consisting of two heterogeneous chaotic maps. Each reference bit triggers the generation of the corresponding chaotic map based on its value. These chaotic sequences form a reference sequence set R; Index modulation and OFDM transmission unit 202, used to transmit the... The index bits are evenly divided into G groups, each group Each set of index bits selects one chaotic sequence from the reference sequence set R through an index selector as the data carrier sequence for that set, resulting in G data carrier sequences. The reference sequence set R and the G data carrier sequences are then superimposed after OFDM modulation to form a transmitted chaotic signal T, which is then sent into the wireless channel. The receiving demodulation unit 203 is used to separate the reference part and the index data part of the received signal after performing cyclic prefix removal, serial-to-parallel conversion, and fast Fourier transform at the receiving end, according to the subcarrier allocation scheme of the transmitting end; input the reference part into a pre-trained deep neural network demodulation model, and output... Reference bits, and the A reference sequence set R is reconstructed from the reference bits according to the mapping relationship; each data-carrying sequence in the index data part is cross-correlated with the reconstructed reference sequence set R, and the sequence position corresponding to the maximum correlation value is used to recover the sequence. Each index bit, thus restoring the entire... One index bit; will one reference bit and The index bits are merged in the order of partitioning to obtain the original transmission bit stream.

[0026] The third embodiment of the present invention provides an OFDM chaotic communication device with reference signal multiplexing, including a memory and a processor. The memory stores a computer program, which can be executed by the processor to implement an OFDM chaotic communication method with reference signal multiplexing as described in any of the above embodiments.

[0027] The fourth embodiment of the present invention provides a computer-readable storage medium, characterized in that it stores a computer program, which can be executed by a processor of the device in which the computer-readable storage medium is located, to implement an OFDM chaotic communication method for reference signal multiplexing as described in any of the above embodiments.

[0028] Based on the OFDM chaotic communication method, apparatus, and medium for reference signal multiplexing provided by this invention, the bit stream to be transmitted is divided into... one reference bit and Each reference bit, with its value of 0 or 1, triggers the corresponding chaotic mapping in a group of chaotic generators composed of two heterogeneous chaotic mappings, generating... A set of chaotic sequences of length β forms a reference sequence set R, which carries both the... The information of each reference bit is used as the codebook for subsequent index modulation; then... Each index bit is uniformly divided into G groups. Each group selects one chaotic sequence from the reference sequence set R as the data carrier sequence using an index selector. The reference sequence set R and the G data carrier sequences are then superimposed after OFDM modulation to form the transmitted signal. At the receiving end, after cyclic prefix removal, serial-to-parallel conversion, and fast Fourier transform, the reference part and the index data part are separated, and the signal is recovered by a deep neural network demodulation model. The reference sequence set R is reconstructed using reference bits, and then the index data is cross-correlation detected and recovered using the reconstructed R. Each index bit enables the reference signal to synchronously complete reference transmission and user information transmission on the same time-frequency and power resources, eliminating the pure overhead of the reference signal.

[0029] Exemplary examples show that the computer program described in the third and fourth embodiments of the present invention can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the OFDM chaotic communication device implementing a reference signal multiplexing. For example, the apparatus described in the second embodiment of the present invention.

[0030] The processor referred to can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the OFDM chaotic communication method for reference signal multiplexing, connecting the various parts of the OFDM chaotic communication method for implementing reference signal multiplexing through various interfaces and lines.

[0031] The memory can be used to store the computer program and / or modules. The processor, by running or executing the computer program and / or modules stored in the memory, and by calling the data stored in the memory, implements various functions of an OFDM chaotic communication method with reference signal multiplexing. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, text conversion function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, text message data, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0032] If the implemented module is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0033] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0034] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An OFDM chaotic communication method with reference signal multiplexing, characterized in that, include: At the transmitting end, the bit stream to be transmitted is divided into... one reference bit and One index bit; the... Each reference bit is sequentially input into a chaos generator group consisting of two heterogeneous chaotic maps. Each reference bit triggers the generation of the corresponding chaotic map based on its value. These chaotic sequences form a reference sequence set R; The The index bits are evenly divided into G groups, each group Each set of index bits selects one chaotic sequence from the reference sequence set R through an index selector as the data carrier sequence for that set, resulting in G data carrier sequences. The reference sequence set R and the G data carrier sequences are then superimposed after OFDM modulation to form a transmitted chaotic signal T, which is then sent into the wireless channel. The receiving end processes the received signal through cyclic prefix removal, serial-to-parallel conversion, and Fast Fourier Transform. Based on the subcarrier allocation scheme of the transmitting end, it separates the reference portion and the index data portion. The reference portion is then input into a pre-trained deep neural network demodulation model, which outputs... Reference bits, and the The reference sequence set R is reconstructed from the reference bits according to the mapping relationship; Each data sequence in the index data portion is cross-correlated with the reconstructed reference sequence set R, and the sequence position corresponding to the maximum correlation value is used to recover that group. Each index bit, thus restoring the entire... One index bit; will one reference bit and The index bits are merged in the order of partitioning to obtain the original transmission bit stream.

2. The OFDM chaotic communication method for reference signal multiplexing according to claim 1, characterized in that, The chaos generator group consisting of two heterogeneous chaotic mappings includes chaos generator A and chaos generator B. Chaos generator A and chaos generator B adopt two independent chaotic mapping methods and have different initial values. When the reference bit is 0, chaos generator A is triggered to generate a chaos generator of length [length missing]. A chaotic sequence, when the reference bit is 1, triggers chaos generator B to generate a sequence of length. A chaotic sequence; The generated after each reference bit is triggered sequentially The chaotic sequences form a reference sequence set in the order of their generation. The reference sequence set R also serves as the input codebook for the index selector.

3. The OFDM chaotic communication method for reference signal multiplexing according to claim 1, characterized in that, The process of each group of index bits selecting one chaotic sequence from the reference sequence set R through the index selector is specifically as follows: For each group of index bits in group G, first... Each bit is converted from binary to its corresponding decimal value; The length generated based on the decimal value is... A binary selection vector, wherein the binary selection vector is 1 only at the position corresponding to the decimal value, and 0 at all other positions; Multiplying the binary selection vector by the reference sequence set R, the chaotic sequence in the reference sequence set R corresponding to the position with a value of 1 is selected as the data carrier sequence for that group. ; The G groups of index bits are processed sequentially to obtain a total of G data-bearing sequences. These sequences are then modulated with the reference sequence set R using OFDM and superimposed to form the transmitted chaotic signal T.

4. The OFDM chaotic communication method for reference signal multiplexing according to claim 1, characterized in that, After generating and transmitting a chaotic signal T into a wireless channel, the signal r(k) received by the receiver satisfies the multipath Rayleigh fading channel model, which contains L independent propagation paths, and the channel response of the l-th path is in complex form: in, This represents the amplitude fading coefficient of the l-th path, which follows a Rayleigh distribution, where j is the imaginary unit. Let l represent the random phase offset of the l-th path and follow a uniform distribution on [0, 2π), where l = 1, 2, ..., L, and L is the maximum number of paths in the channel; The received signal r(k) obtained by the receiver consists of a multipath delay superposition signal and additive noise, and its mathematical model is as follows: in Delay the transmitted signal along the l-th path The value at the kth sampling point after sampling points. The propagation delay of the l-th path, With a mean of 0 and a variance of Additive white Gaussian noise, This represents the one-sided noise power spectral density.

5. The OFDM chaotic communication method for reference signal multiplexing according to claim 1, characterized in that, The process involves performing cross-correlation operations between each data-carrying sequence in the index data portion and the reconstructed reference sequence set R, and then recovering the sequence from the position corresponding to the maximum correlation value. The index bits are specifically: Will The reference sequence set R, reconstructed from the reference bits according to the mapping relationship, is constructed as a reference matrix. The j-th row of the reference matrix B is the j-th chaotic sequence in the reference sequence set R; For each data sequence in the index data section, construct a data matrix. The data matrix A and the reference matrix B are then cross-correlated using the following formula: The length is obtained as The relevant value vector Z; Find the position corresponding to the maximum absolute value in each component of the relevant value vector Z, and restore the decimal value corresponding to that position by performing the inverse mapping from binary to decimal. Each index bit is used as the demodulation result for this group; The data carrying sequences of group G are processed sequentially and concatenated in group order to obtain the complete sequence. Index bits.

6. The OFDM chaotic communication method for reference signal multiplexing according to claim 1, characterized in that, The deep neural network demodulation model consists of an input layer, four one-dimensional convolutional layers, a self-attention layer, two bidirectional LSTM layers, two fully connected layers with random deactivation, and a Softmax output layer connected in sequence. Its demodulation process for the reference part is as follows: The chaotic sequence of the reference part is Z-score standardized to make its mean 0 and variance 1; The standardized sequence is sequentially input into four one-dimensional convolutional layers. Each convolution operation is followed by a ReLU activation function and layer normalization. The number of output channels and the width of the convolutional kernel increase layer by layer, and the output local feature matrix H is generated. The local feature matrix H is input into the attention layer, and the attention weights are calculated using the learnable query vector Q to obtain the weighted feature matrix. ; The weighted feature matrix Input a bidirectional LSTM layer, which contains two layers of bidirectional LSTM units with 128 hidden units each, and output global temporal features; The global temporal features and the local feature matrix H are concatenated and then sequentially fed into two fully connected layers with random deactivation and a Softmax output layer to output the probability vector of the nth reference bit. ,in and These represent the probabilities that the reference bit is 0 and 1, respectively. For the probability vector Make the highest probability decision, and The binary symbol corresponding to the one with the higher probability value is used as the decision result for that reference bit, and this is applied sequentially to all bits. The decision is made using reference bits to obtain... One reference bit.

7. The OFDM chaotic communication method for reference signal multiplexing according to claim 1, characterized in that, The deep neural network demodulation model is obtained through offline training, and the offline training process includes: For each training reference bit, depending on whether its value is 0 or 1, either chaos generator A or chaos generator B is invoked to generate a chaotic sequence of length β as a reference sequence. After simulated transmission through a channel within a set signal-to-noise ratio range, M training sample pairs are formed. ,in The received subsequence after transmission through the channel. For the corresponding binary tag; The M training samples are divided into a training set and a validation set according to the set validation set ratio, and the batch size, learning rate and maximum number of training rounds are set. The difference between the predicted output and the true label of a deep neural network demodulation model is evaluated using binary cross-entropy as the loss function. The loss function is defined as follows: Where M is the number of training samples. Let i be the true label of the i-th training sample. This is the estimated output of the deep neural network demodulation model for the i-th training sample; An adaptive moment estimation optimizer is used to perform multiple rounds of iterative optimization on the parameters of the deep neural network demodulation model through a three-stage loop: forward inference to calculate the predicted distribution, backpropagation to obtain the gradient from the loss function, and parameter update. After each training cycle, the generalization loss of the model is evaluated on the validation set. Training is terminated when the validation loss does not decrease significantly within a set number of consecutive training cycles, and the model with the best validation performance is saved as the deep neural network demodulation model.

8. An OFDM chaotic communication device with reference signal multiplexing, characterized in that, include: The reference sequence generation unit is used at the transmitting end to divide the bit stream to be transmitted into... one reference bit and One index bit; the... Each reference bit is sequentially input into a chaos generator group consisting of two heterogeneous chaotic maps. Each reference bit triggers the generation of the corresponding chaotic map based on its value. These chaotic sequences form a reference sequence set R; Indexed modulation and OFDM transmission unit, used to transmit the... The index bits are evenly divided into G groups, each group Each set of index bits selects one chaotic sequence from the reference sequence set R through an index selector as the data carrier sequence for that set, resulting in G data carrier sequences. The reference sequence set R and the G data carrier sequences are then superimposed after OFDM modulation to form a transmitted chaotic signal T, which is then sent into the wireless channel. The receiving demodulation unit is used to separate the reference part and the index data part of the received signal after performing cyclic prefix removal, serial-to-parallel conversion, and fast Fourier transform at the receiving end, according to the subcarrier allocation scheme at the transmitting end; the reference part is input into a pre-trained deep neural network demodulation model, and the output is... Reference bits, and the The reference sequence set R is reconstructed from the reference bits according to the mapping relationship; Each data sequence in the index data portion is cross-correlated with the reconstructed reference sequence set R, and the sequence position corresponding to the maximum correlation value is used to recover that group. Each index bit, thus restoring the entire... One index bit; will one reference bit and The index bits are merged in the order of partitioning to obtain the original transmission bit stream.

9. An OFDM chaotic communication device with reference signal multiplexing, characterized in that, The system includes a memory and a processor. The memory stores a computer program that can be executed by the processor to implement an OFDM chaotic communication method for reference signal multiplexing as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The device contains a computer program that can be executed by a processor of the device in which the computer-readable storage medium is located, to implement an OFDM chaotic communication method for reference signal multiplexing as described in any one of claims 1 to 7.

Images