Satellite-ground network OFDM anti-interception waveform design method based on security coding

Abstract

The invention provides a satellite-ground network OFDM (Orthogonal Frequency Division Multiplexing) anti-interception waveform design method based on security coding, and relates to the technical field of communication. The method comprises the following steps: determining a first signal to interference plus noise ratio at a legal receiver and a second signal to interference plus noise ratio at an eavesdropper according to a source symbol transmitted on each subcarrier, a security coding matrix in a time domain, a power allocation scheme, a frequency domain channel matrix and a frequency domain noise vector; according to the first signal to interference plus noise ratio and the second signal to interference plus noise ratio, a target optimization problem is established, and the target optimization problem takes the signal quality of the legal receiver end as an optimization target; and solving the target optimization problem by using a binary search algorithm and a neural network to obtain a final security coding matrix and a final power allocation scheme, and designing a satellite-ground network OFDM anti-interception waveform based on security coding by using the final security coding matrix and the final power allocation scheme. Therefore, the secrecy performance of the satellite-to-ground link is relatively high, and the computing power resource occupation of the satellite-to-ground fusion network is reduced.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a satellite-to-ground network OFDM anti-interception waveform design method based on security coding. Background Art

[0002] In the past few years, traditional terrestrial wireless communications have experienced rapid growth in both user numbers and service demand. Future networks will need to provide more resources than current networks. Limited by communication capacity and coverage, terrestrial networks alone cannot meet these demands. Low Earth Orbit (LEO) satellites, however, have regained public attention due to their wide coverage, low latency, minimal loss, strong anti-interference capabilities, and all-weather, all-day operation. Currently, LEO satellite systems generally utilize orthogonal frequency division multiplexing (OFDM) technology, which is widely used for its high spectral efficiency and high data rates. However, despite significant advances in communication technology, the inherent vulnerability of wireless channels poses significant security risks to satellite-to-ground communications. Due to the broadcast nature of wireless communications, satellite-to-ground links are highly vulnerable to both passive and active attackers (primarily eavesdroppers). Designing secure communication algorithms based on the characteristics of existing satellite-to-ground networks to achieve downlink security has become a major challenge facing these networks.

[0003] At present, the key-based physical layer security technology in existing technologies may have key unification problems, which will lead to information leakage and reduce the confidentiality performance of the satellite-to-ground link. The satellite-to-ground fusion network environment changes rapidly, and on-board resources are limited. Traditional optimization algorithms lack generalization and occupy more computing power resources of the satellite-to-ground fusion network. Summary of the Invention

[0004] The purpose of the embodiments of the present invention is to provide a satellite-to-ground network OFDM anti-interception waveform design method based on security coding, which solves the problems of low confidentiality performance of satellite-to-ground links and more computing power resources occupied by traditional optimization algorithms in the satellite-to-ground fusion network.

[0005] To solve the above technical problems, the embodiments of the present invention provide the following technical solutions:

[0006] A first aspect of the present invention provides a satellite-to-ground network OFDM anti-interception waveform design method based on security coding, comprising:

[0007] Determine a first signal-to-interference-and-noise ratio (SINR) at a legitimate receiver and a second SINR at an eavesdropper based on source symbols transmitted on each subcarrier, a security coding matrix in the time domain, a power allocation scheme, a frequency domain channel matrix, and a frequency domain noise vector;

[0008] Establishing a target optimization problem based on the first signal to interference plus noise ratio and the second signal to interference plus noise ratio, wherein the target optimization problem takes the signal quality of the legitimate receiver end as an optimization target;

[0009] The binary search algorithm and neural network are used to solve the target optimization problem and obtain the final security coding matrix and the final power allocation scheme. The final security coding matrix and the final power allocation scheme are used to design the OFDM anti-interception waveform of the satellite-to-ground network based on security coding.

[0010] A second aspect of the present invention provides a satellite-to-ground network OFDM anti-interception waveform design device based on security coding, comprising:

[0011] a determination module, configured to determine a first signal-to-interference-and-noise ratio (SINR) at a legitimate receiver and a second SINR at an eavesdropper based on source symbols transmitted on each subcarrier, a security coding matrix in the time domain, a power allocation scheme, a frequency domain channel matrix, and a frequency domain noise vector;

[0012] An establishing module, configured to establish a target optimization problem based on the first signal to interference plus noise ratio and the second signal to interference plus noise ratio, wherein the target optimization problem takes the signal quality of the legitimate receiver end as an optimization target;

[0013] The solution module is used to solve the target optimization problem using a binary search algorithm and a neural network to obtain the final security coding matrix and the final power allocation scheme. The final security coding matrix and the final power allocation scheme are then used to design an OFDM anti-interception waveform for the satellite-to-ground network based on security coding.

[0014] Compared to the prior art, the present invention provides a method for designing an OFDM waveform for satellite-to-ground networks based on secure coding. This method determines a first signal-to-interference-and-noise ratio (SINR) at the legitimate receiver and a second SINR at the eavesdropper based on the source symbols transmitted on each subcarrier, a time-domain security coding matrix, a power allocation scheme, a frequency-domain channel matrix, and a frequency-domain noise vector. Based on the first and second SINRs, a target optimization problem is established, with the signal quality at the legitimate receiver as the optimization objective. This target optimization problem is solved using a binary search algorithm and a neural network to obtain a final security coding matrix and a final power allocation scheme. These final security coding matrix and power allocation scheme are then used to design an OFDM waveform for satellite-to-ground networks based on secure coding. This target optimization problem, which optimizes the signal quality at the legitimate receiver and utilizes the time-domain security coding matrix, improves the confidentiality of the satellite-to-ground link. The binary search algorithm and low-complexity neural network can reduce computing resource usage in the satellite-to-ground converged network. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0016] Figure 1 The flowchart of the OFDM anti-interception waveform design method for satellite-to-ground network based on security coding is schematically shown;

[0017] Figure 2 Schematically shows an OFDM system scenario diagram;

[0018] Figure 3 The flowchart of the algorithm for solving the target optimization problem by using the binary search algorithm and the neural network is schematically shown;

[0019] Figure 4 The relationship between the total transmission power and the signal-to-interference-and-noise ratio of legitimate users is schematically shown;

[0020] Figure 5 The relationship between the total transmission power and the signal-to-interference-and-noise ratio of the eavesdropper is schematically shown;

[0021] Figure 6 The relationship between the total transmission power and the confidentiality rate is schematically shown;

[0022] Figure 7 The comparative experiment confidentiality rate performance diagram is schematically shown;

[0023] Figure 8 The relationship between symbol error rate and confidentiality rate performance is schematically shown;

[0024] Figure 9 The structure of the OFDM anti-interception waveform design device for satellite-to-ground network based on security coding is schematically shown. DETAILED DESCRIPTION

[0025] Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0026] It should be noted that, unless otherwise specified, the technical or scientific terms used in the present invention should have the common meanings understood by those skilled in the art to which the present invention belongs.

[0027] The method in the embodiment of the present invention is described in detail below.

[0028] Figure 1 The flowchart of the OFDM anti-interception waveform design method for satellite-to-ground network based on security coding in an embodiment of the present invention is schematically shown. Figure 1 As shown, the satellite-to-ground network OFDM anti-interception waveform design method based on security coding may include:

[0029] This paper studies secure satellite-to-ground downlink communications in remote areas with sparsely distributed users. This study focuses on the worst-case scenario, assuming that the channel between the satellite and legitimate users has similar propagation characteristics to the channel between the satellite and an eavesdropper. Under this scenario, from an information-theoretic security perspective, relying on the inherent randomness of the channel for secure transmission is difficult. This is particularly true in widely used OFDM systems, where the orthogonality between subcarriers leads to high correlation between the subchannels between the satellite and the ground, further exacerbating the difficulties of secure transmission. Figure 2 The OFDM system scenario diagram is schematically shown, see Figure 2 As shown in Figure 1, the OFDM system uses Fast Fourier Transform (FFT) and Inverse Fast Fourier Transform (IFFT) for signal processing. At the transmitter, the input bit stream is first generated into complex frequency domain symbols through constellation mapping (such as QPSK). These symbols are then input into an N-point IFFT module, which converts them into time domain signals by superimposing N orthogonal subcarriers. The amplitude and phase information of each complex symbol determines the amplitude and phase of its corresponding subcarrier. The time domain signal is the superposition result of all modulated subcarriers.

[0030] S101. Determine a first signal-to-interference-plus-noise ratio (SINR) at a legitimate receiver and a second SINR at an eavesdropper based on source symbols transmitted on each subcarrier, a security coding matrix in the time domain, a power allocation scheme, a frequency domain channel matrix, and a frequency domain noise vector.

[0031] Assume that the source symbol transmitted on each OFDM subcarrier can be expressed as:

[0032]

[0033] Among them, S represents the source symbol transmitted on the subcarrier, S k represents the source symbol transmitted on the kth subcarrier, |S k | 2 =1 means the Sth k The power of the symbol in the subchannel corresponding to the subcarriers is N, where N represents the number of subcarriers. Typically, a receiver receives a signal transmitted via a wireless channel and converts it to the frequency domain using FFT to recover the data.

[0034] The satellite-to-ground communication channel is modeled as follows:

[0035]

[0036] Among them, C L represents the free space path loss, which is expressed as C L =(λ / 4π) 2 / (d 2 +l 2 ), where λ is the signal wavelength, d is the horizontal distance between the satellite projection point and the ground user, l is the satellite's orbital altitude, and θ is the phase offset term. The parameter β represents the fading caused by rainfall and obeys the log-normal distribution, that is, μ represents the mean, δ 2 represents the variance, where β dB =10log 10 β. The phase shift term θ obeys a uniform distribution in the interval [0,2π).

[0037] The beam gain b is defined as:

[0038]

[0039] Where G0 is the maximum gain of the antenna, u0=2.07123sin(α) / sin(α 3dB ), α represents the elevation angle between the beam center and the user, α 3dB is the 3 dB beamwidth of the antenna, J1(·) and J3(·) are the first-order and third-order Bessel functions of the first kind, respectively.

[0040] To mitigate the potential eavesdropping threat in satellite-to-ground OFDM communication systems, this paper proposes a satellite-to-ground network OFDM waveform design method based on secure coding. Taking into account the strong channel similarity between legitimate and eavesdropped links, this method introduces secure coding before wireless transmission and enhances confidentiality by intentionally introducing inter-subcarrier interference. Specifically, this paper designs a pre-transmission signal processing technique that introduces controllable waveform aliasing between subchannels to interfere with potential eavesdroppers' recognition of transmitted symbols. Subsequently, the encoded transmitted signal x is expressed as:

[0041]

[0042] Among them, m represents the security coding matrix in the time domain, represents a set of N×N complex matrices, represents the circular convolution operator, ⊙ represents the Hadamard product, F is the normalized FFT matrix, p is the power allocation scheme, in, Represents the square root of the transmission power allocated to the kth subcarrier.

[0043] We can further obtain the frequency domain signal at the receiving end:

[0044] Y b =HX+N b ;

[0045] Y e =GX+N e ;

[0046] Among them, Y b Represented as the frequency domain noise vector received by the legitimate user, Y e represents the frequency domain noise vector received by the eavesdropper, H is the frequency domain channel matrix from the legitimate transmitter to the legitimate receiver, G is the frequency domain channel matrix from the legitimate transmitter to the eavesdropper, and H and G are both diagonal matrices. Based on the expression of the source symbol transmitted on each OFDM subcarrier and the expression of the encoded transmitted signal, the encrypted symbol is determined. The encrypted symbol X can be expressed as:

[0047]

[0048] in, is the square root of the transmission power allocated to the first subcarrier, n is the nth subcarrier, N is the number of subcarriers, M 1,n is the element of the 1st row and nth column of the security coding matrix in the frequency domain, S n is the source symbol transmitted on the nth subcarrier, The square root of the transmission power allocated to the nth subcarrier, M N,n is the element in the Nth row and nth column of the security coding matrix in the frequency domain.

[0049] The signal model of the OFDM anti-interception waveform at the receiving end will be reconstructed. Therefore, the signal received by the legitimate receiver from the subchannel corresponding to the kth subcarrier is It can be expressed as:

[0050]

[0051] Among them, H k represents the k-th diagonal element of the frequency domain channel matrix H from the legal transmitter to the legal receiver, corresponding to the channel gain of the k-th subcarrier, represents the frequency domain noise on the kth subcarrier received by the legitimate receiver, M k,n is the element of the kth row and nth column of the security coding matrix in the frequency domain, M k,k is the k-th element in the k-th row of the security coding matrix M in the frequency domain.

[0052] The signal received by the subchannel corresponding to the kth subcarrier at the eavesdropper is It can be expressed as:

[0053]

[0054] Among them, G k represents the k-th diagonal element of the frequency domain channel matrix G from the legitimate transmitter to the eavesdropper, corresponding to the channel gain of the k-th subcarrier, represents the frequency domain noise on the kth subcarrier received by the eavesdropper.

[0055] according to and The expression of , determines the first signal to interference and noise ratio at the legitimate receiver and the second signal-to-interference-and-noise ratio at the eavesdropper

[0056]

[0057] in, represents the noise power of the legal subchannel corresponding to the kth subcarrier, is the noise power on the eavesdropping subchannel corresponding to the kth subcarrier, p k is the transmission power of the kth subcarrier.

[0058] S102: Establish a target optimization problem according to the first signal to interference plus noise ratio and the second signal to interference plus noise ratio.

[0059] Among them, the target optimization problem takes the signal quality of the legitimate receiver as the optimization target.

[0060] Based on the first signal-to-interference-and-noise ratio and the second signal-to-interference-and-noise ratio, physical layer security coding is required to establish a controlled asymmetric signal quality distortion between the legitimate user and the eavesdropper. Specifically, it is necessary to ensure that the signal quality of the legitimate user is not severely distorted while reducing the signal quality at the eavesdropper. The key lies in how to balance the conflict between reliability and security. To this end, the security coding matrix designed by the present invention aims to fully utilize the difference in channel state information between the legitimate and eavesdropped channels, and formally express M=ψ(h b ,h e ), where ψ(·) represents the feature extraction function, h b is the time domain channel matrix from the legal transmitter to the legal receiver, h e is the time domain channel matrix from the legitimate transmitter to the eavesdropper. The reliability performance of the OFDM system can be quantified by the Symbol Error Rate (SER).

[0061] Specifically, based on the first signal to interference plus noise ratio and the second signal to interference plus noise ratio, a target optimization problem is established, including:

[0062] Step A1: performing an open operation on an average value of the first signal to interference plus noise ratio and the second signal to interference plus noise ratio to obtain a symbol error rate.

[0063] Symbol Error Rate (SER) of OFDM Systems Modulated with Quadrature Phase Shift Keying (QPSK) The expression is:

[0064]

[0065] in, represents the average signal to interference and noise ratio, that is, the average of the first signal to interference and noise ratio and the second signal to interference and noise ratio, and erfc(·) represents the complementary error function.

[0066] Step A2: Establish a first optimization problem based on the first signal to interference plus noise ratio and the second signal to interference plus noise ratio.

[0067] The first optimization problem takes maximizing the confidentiality rate as an optimization goal, and the first optimization problem satisfies the constraint of the symbol error rate.

[0068] Specifically, step A2 includes:

[0069] Step A21: Determine a confidentiality rate according to the first signal to interference plus noise ratio and the second signal to interference plus noise ratio.

[0070] The reliability of OFDM system can be measured by the confidentiality rate, R k The expression is:

[0071]

[0072] Step A22: Establish a first optimization problem based on the confidentiality rate, the transmission power of each subcarrier of the eavesdropper, the error threshold of the eavesdropper, the transmission power of each subcarrier of the legitimate receiver, the error threshold of the legitimate receiver, the transmission power of the subcarrier, the security coding matrix in the frequency domain, and the transmission power upper limit of each subcarrier.

[0073] To address the key reliability-security trade-off issue, the present invention proposes a confidentiality rate maximization optimization model, namely the first optimization problem, by generating a secure coding matrix and optimizing the power allocation of each subcarrier while satisfying the symbol error rate constraints of legitimate users and eavesdroppers.

[0074] Specifically, the expression of the first optimization problem is:

[0075]

[0076] in, is the first optimization problem, R k is the confidentiality rate of the kth subcarrier, M is the security coding matrix in the frequency domain, p is the power allocation scheme, P b is the transmit power of each subcarrier of the legal receiver, ε b is the error threshold of the legitimate receiver, P e is the eavesdropper’s transmission power of each subcarrier, ε e is the error threshold of the eavesdropper, p k is the transmission power of the kth subcarrier, n is the nth subcarrier, N is the number of subcarriers, M k,n is the element of the kth row and nth column of the security coding matrix in the frequency domain, P S The upper limit of the transmit power for each subcarrier.

[0077] The first optimization problem is to maximize the confidentiality rate and optimize the security coding matrix M and power allocation scheme p in the frequency domain, where P b ≤ε b and P e ≥ε e It is an anti-eavesdropping constraint, Represents the power constraint for each subcarrier.

[0078] Step A3: reconstruct the first optimization problem according to the first signal to interference and noise ratio, the symbol error rate, the transmit power of each subcarrier of the eavesdropper, and the error threshold of the eavesdropper to establish a target optimization problem.

[0079] Among them, the target optimization problem takes the signal quality of the legitimate receiver as the optimization target.

[0080] In order to solve the non-convex problem, that is, the first optimization problem Combined with formula P e ≥ε e , the first optimization problem can be reconstructed into a maximization problem with the goal of improving the signal quality of the legitimate receiver. Symbol Error Rate and P b ≤ε b It can be seen that the communication rate at the legitimate receiver and its symbol error rate Therefore, the constraint P b ≤ε b Relaxation is performed to reformulate the first optimization problem into the target optimization problem.

[0081] Specifically, the expression of the target optimization problem is:

[0082]

[0083] in, is the target optimization problem, M is the security coding matrix in the frequency domain, p is the power allocation scheme, is the first signal to interference and noise ratio of the kth subcarrier at the legal receiver, P e is the eavesdropper’s transmission power of each subcarrier, ε e is the error threshold of the eavesdropper, p k is the transmission power of the kth subcarrier, n is the nth subcarrier, N is the number of subcarriers, M k,n is the element of the kth row and nth column of the security coding matrix in the frequency domain, P S The upper limit of the transmit power for each subcarrier.

[0084] S103. Solve the target optimization problem using a binary search algorithm and a neural network to obtain a final security coding matrix and a final power allocation scheme. Use the final security coding matrix and the final power allocation scheme to design a satellite-to-ground network OFDM anti-interception waveform based on security coding.

[0085] Specifically, Figure 3 The flowchart of the algorithm for solving the target optimization problem using the binary search algorithm and the neural network is shown schematically. Figure 3 As shown, the target optimization problem is The solution is broken down into two sub-problems: In the first phase, the subcarrier power allocation is fixed to generate the corresponding final security coding matrix. In the second phase, the subcarrier power allocation strategy is further optimized based on the final security coding matrix. These two phases are then iterated alternately until system performance converges.

[0086] Step S103 includes:

[0087] Step B1: Decompose the target optimization problem into a secure coding optimization problem.

[0088] Specifically, step B1 includes: setting the power allocation scheme to a fixed value to decompose the target optimization problem into a secure coding optimization problem.

[0089] Among them, the security coding optimization problem takes the signal quality of the legitimate receiver as the optimization goal.

[0090] The expression of the secure coding optimization problem is:

[0091]

[0092] in, is a security coding optimization problem, M is a security coding matrix in the frequency domain, is the first signal to interference and noise ratio of the kth subcarrier at the legal receiver, M k,nis the element of the kth row and nth column of the security coding matrix in the frequency domain, P S The transmit power upper limit of each subcarrier, n is the nth subcarrier, and N is the number of subcarriers.

[0093] Step B2: Input the composite channel state matrix into the complex multilayer perceptron of the neural network to simulate an equivalent complex transformation through the real-valued fully connected layer of the complex-valued multilayer perceptron (CMLP), and output an optimized security coding matrix in the frequency domain.

[0094] Specifically, in order to solve the non-convex problem, namely the secure coding optimization problem The present invention designs a complex multilayer perceptron that can directly process the native complex representation of the legitimate channel and the eavesdropping channel parameters. The input of the neural network is the composite channel state matrix H′, H′=H′ real +iH′ imag , where H′ real represents the real component of the composite channel state matrix, H′ imag represents the imaginary component of the composite channel state matrix, and i is the imaginary unit. The implementation of CMLP uses paired real-valued fully connected layers to simulate complex-valued transformations. For any complex input vector, the inter-layer mapping of CMLP follows the following complex linear mapping relationship:

[0095] x real =Lin(H′ real )-Lin(H′ imag );

[0096] x imag =Lin(H′ imag )+Lin(H′ real );

[0097] Among them, Lin(·) represents a real-valued fully connected layer, which performs an affine transformation on the input, that is, Lin(x) = Wx + b, where W represents the learnable weight parameter and b represents the bias vector. In order to preserve the inherent amplitude and phase correlation in complex data, CMLP simulates the equivalent complex transformation through a traditional real-valued fully connected layer. Three complex-valued fully connected layers are used in the neural network, and their output dimensions are set to 128, 64, and 16 respectively. The LeakyReLU activation function is introduced in each layer to achieve nonlinear transformation. The mathematical expression of the LeakyReLU activation function is:

[0098]

[0099] Among them, σ′ is a small constant (usually a value such as 0.01 or 0.2), which is used to maintain a non-zero gradient when the input is negative, thereby effectively alleviating the gradient vanishing problem.

[0100] The Adam algorithm is used to train the neural network. This algorithm dynamically adjusts the learning rate based on the first- and second-order moment estimates of the gradient, achieving robust and efficient convergence. Ultimately, the neural network outputs a frequency-domain optimized secure coding matrix for the OFDM system.

[0101] The update rule of the neural network parameter θ′ in the tth iteration is as follows:

[0102]

[0103] Among them, θ t ′ is the neural network parameter of the tth iteration, θ′ t-1 is the neural network parameter of the t-1th iteration, ∈ is a very small constant used to prevent division by zero errors, α is the initial learning rate, is the bias-corrected estimate of the first moment of the gradient, is the bias-corrected estimate of the second moment of the gradient, and Defined as:

[0104]

[0105] in:

[0106]

[0107] Among them, β1 is the exponential decay rate of the momentum of the first-order moment (mean) of the control gradient, and β2 is the exponential decay rate of the momentum of the second-order moment (variance). Represents the loss gradient of the tth round.

[0108] Step B3: Using an activation function to process the security coding matrix optimized in the frequency domain, to obtain an activation function of the security coding matrix optimized in the frequency domain.

[0109] Specifically, for constraints Introduce a designed activation function, denoted as Used to ensure that the output of the neural network meets the power constraint. This activation function acts on a complex-valued vector to ensure that the square norm of the complex-valued vector does not exceed the given power threshold P S Activation function The definition is as follows:

[0110]

[0111] Among them, M krepresents the kth row of the optimized security coding matrix in the frequency domain, including all column elements of the kth row. k If the square norm of M is within the limit, then keep M k Output the original value of M; otherwise, scale it proportionally. k Projection into the feasible region. This projection activation strategy not only preserves the ability of the neural network to approximate the optimal solution in non-convex space, but also ensures that the neural network output strictly meets the power constraints of the system physical layer, thus providing feasibility guarantees for subsequent transmission processes.

[0112] Introduction After the activation function, the loss gradient of the tth round is defined as follows:

[0113]

[0114] The loss gradient of the tth round In this way, the neural network can achieve the training goal of unsupervised learning through the back propagation mechanism under the premise of meeting the power constraint.

[0115] Step B4: According to the ratio of the security coding matrix optimized in the frequency domain to the activation function of the security coding matrix optimized in the frequency domain, the security coding optimization problem is solved to determine the final security coding matrix.

[0116] Secure Coding Optimization Problem The final security encoding matrix M k The expression of ′ is:

[0117]

[0118] Through the activation function Constraints can be satisfied The final security coding matrix.

[0119] Step B5: Decompose the target optimization problem into a power allocation optimization problem.

[0120] Specifically, step B5 includes: decomposing the target optimization problem into a power allocation optimization problem according to the transmission power of the subcarrier and the minimum subcarrier symbol error rate.

[0121] Among them, the power allocation optimization problem takes the transmission power of the subcarrier as the objective function.

[0122] Based on the final security coding matrix, the power allocation scheme is further optimized. Based on the target optimization problem, the anti-eavesdropping constraint P e ≥ε e It can be equivalently expressed as the symbol error rate of the minimum subcarrier meeting the anti-eavesdropping constraint, while the power constraint Transformed into Since the optimized security coding matrix in the frequency domain obtained by neural network optimization can maximize the system confidentiality rate, the power allocation scheme maximizes the system confidentiality rate when it is close to the original allocation, that is, maximizes the total allocated power optimization goal.

[0123] Therefore, the expression of the power allocation optimization problem is:

[0124]

[0125] in, is the power allocation optimization problem, p k is the transmission power of the kth subcarrier, is the symbol error rate of the first signal to interference and noise ratio of the kth subcarrier at the legal receiver, ε e is the error threshold of the eavesdropper, n is the nth subcarrier, N is the number of subcarriers, M k,n is the element of the kth row and nth column of the security coding matrix in the frequency domain, P S The upper limit of the transmit power for each subcarrier.

[0126] Step B6: Use the binary search algorithm to solve the power allocation optimization problem and obtain the final power allocation scheme. Use the final security coding matrix and the final power allocation scheme to design a satellite-to-ground network OFDM anti-interception waveform based on security coding.

[0127] In order to solve the power allocation optimization problem Using the binary search algorithm, an iterative power allocation optimization algorithm inspired by the binary search algorithm is proposed to solve the power allocation scheme. The iterative power allocation optimization algorithm is as follows:

[0128] Step B61: Initialization and p min =0.

[0129] Step B62: According to , calculate the symbol error rate of each subchannel on the eavesdropper side, determine the subchannel corresponding to the minimum symbol error rate, and mark it with index index, that is,

[0130] tolerance limit, execute step B64 to update the index power.

[0131] Step B64: Calculate the intermediate power value p based on the binary search algorithm mid , p mid =(p min +p max ) / 2.

[0132] Step B65: According to The expression and intermediate power value p mid Calculation of symbol errors

[0133] Step B66: If Update p min =p mid , go to step B64; if Update p max =p mid , execute step B64.

[0134] Step B67: If That is, the symbol error rate meets the constraint, the index power is retained and the cycle is stopped, and the final power allocation solution is obtained.

[0135] Among them, p min is the minimum power value, p max is the maximum power value, M′ k,n is the element in the kth row and nth column of the security coding matrix in the frequency domain of the final security coding matrix.

[0136] Based on the final power allocation solution output by the above iterative power allocation optimization algorithm, Can be further updated to:

[0137]

[0138] in, is the updated loss gradient of the tth round, M′ k,k is the element of the k-th row and k-th column of the security coding matrix in the frequency domain of the final security coding matrix, is the noise power received by the kth subcarrier.

[0139] The updated loss function incorporates the optimized power allocation scheme and is used during the back-propagation process in the training phase to update the weights and parameters of the neural network in secure coding optimization.

[0140] In the inference stage, after the neural network optimizes the security coding matrix, it optimizes the power allocation scheme based on the iterative power allocation optimization algorithm. The security coding matrix and power allocation scheme are the final optimization scheme.

[0141] To verify the reliability and effectiveness of the present invention, a variety of simulations were conducted to evaluate the confidentiality performance of the proposed scheme. The simulation parameters were set as follows: the satellite transmitter operated at an orbital altitude of 600 kilometers, and the ground base stations were randomly distributed within a radius of 800 kilometers centered on the center of the coverage area. The eavesdropper was also located on the ground, with its coordinates randomly distributed within a radius of 1000 kilometers centered on the legitimate satellite transmitter. At a reference distance of 1 meter, the channel power gains from the satellite to the legitimate ground receiver and from the satellite to the ground eavesdropper were set to 10dB and 5dB, respectively. For the downlink scenario, the propagation channel between the satellite and the ground receiver adopts the Rician fading model, with a Rician factor set to 10dB.

[0142] First, we explored the impact of transmit power on the SINR performance of the legal receiving end. The experimental results are as follows: Figure 4 As shown, Figure 4 The diagram schematically illustrates the relationship between total transmit power and the Signal-to-Interference-Noise Ratio (SINR) of legitimate users. As the power constraint for each subcarrier increases, the SINR improves accordingly. This is due to the enhanced primary channel quality under the anti-interception constraints. However, after a certain power level, the SINR growth rate flattens. This can be attributed to two factors: 1. The noise-limited region: At lower power levels, ambient noise dominates, limiting the symbol error rate (SER) of eavesdroppers. In this region, available power is primarily used to improve the signal quality of legitimate users. 2. The interference allocation mechanism: As power increases, more energy is allocated to inter-subcarrier interference (ISI) to reduce the SER of eavesdroppers, leading to saturation of the SINR gain for legitimate users. There is an inherent trade-off between confidentiality and performance in the system: stricter ISI constraints (i.e., lower SINR for eavesdroppers) reduce the SINR for legitimate users because the system must allocate more power to ISI generation. When the eavesdropper's symbol error rate reaches 0.5 (i.e., no valid information can be extracted), the effective information rate in the legitimate channel will also collapse - this is a performance degradation that is actively introduced for security purposes.

[0143] Figure 5 The relationship between total transmit power and the eavesdropper's signal-to-interference-and-noise ratio (SINR) is schematically shown. The SINR performance of the eavesdropper under different anti-interception constraints is tested. It can be seen that regardless of how the upper limit of the transmission power is changed, the SINR performance of the eavesdropper remains within a certain range. Calculations show that the corresponding symbol error rate performance of the eavesdropper is highly consistent with the set anti-interception constraints. This verifies the effectiveness of the proposed two-stage optimization method in optimizing the power allocation scheme, limiting the eavesdropper's SINR performance, and satisfying the anti-interception constraints.

[0144] Figure 6The relationship between total transmit power and confidentiality rate is schematically shown, and the impact of the subcarrier transmission power cap on the system's confidentiality rate performance is tested. It is clear that with increasing transmission power cap, confidentiality rate performance shows an upward trend, which is consistent with the change in SINR performance of both the legitimate user and eavesdropper as the transmission power cap is increased. When the anti-interception constraint is set to 0.4, the system can effectively achieve a high confidential communication rate while ensuring anti-interception performance. However, when the SER is set to 0.5, although the eavesdropper is completely interfered with, the legitimate user cannot obtain any useful information, causing the confidentiality rate performance to drop to 0.

[0145] Figure 7 A comparative experiment shows a confidentiality rate performance diagram, testing different transmission schemes as the confidentiality rate performance varies with the maximum transmission power constraint. The results show that the confidentiality rate performance is significantly lower in the unencrypted scheme and the Maximum Ratio Transmission (MRT) scheme. This is because the performance of these two schemes primarily relies on channel state information (CSI). Even if the transmission power limit is increased, the SINR performance of both legitimate users and eavesdroppers will increase, thus limiting the improvement in confidentiality rate. In contrast, the proposed method significantly improves confidentiality rate performance. Furthermore, it can be observed that the greater the number of subcarriers, the better the system's confidentiality rate performance. This trend can be explained analytically: more subcarriers increase the denominator in the formula, meaning that with a larger number of subcarriers, the required inter-subcarrier interference can be relatively low to achieve the same anti-eavesdropping effect, as the denominator is composed of the sum of inter-subcarrier interference and ambient noise.

[0146] Figure 8 The relationship between symbol error rate and confidentiality rate performance is shown schematically in FIG. Figure 8 As shown in Figure 2, the impact of anti-interception constraints on the system's confidentiality rate performance was tested. As the anti-interception constraints increased, the confidentiality rate performance gradually decreased. This is because higher anti-interception constraints require more power to be allocated for interference to prevent eavesdropping. Under the same transmission power limit, this will lead to a decrease in the SINR performance of the legitimate user end, and thus a decrease in the confidentiality rate performance. On the other hand, increasing the number of subcarriers will improve the confidentiality rate performance. This is because when there are more subcarriers, the interference power required on each subcarrier is relatively small, which improves the SINR performance of each subcarrier for legitimate users and ultimately improves the overall confidentiality rate. By combining the characteristics of OFDM modulation with the principles of physical layer security, the method proposed in this article achieves a good balance between power efficiency and security requirements.

[0147] The present invention targets satellite-ground fusion networks with sparsely distributed users in remote areas, and uses a secure coding mechanism to actively introduce inter-subcarrier interference. While achieving physical layer secure transmission, it can significantly reduce the cost of achieving secure transmission in remote areas. In order to simplify the problem of maximizing the system confidentiality rate, that is, the target optimization problem, the target optimization problem is broken down into two sub-problems for solution. The unsupervised learning optimization of the secure coding scheme can significantly reduce the algorithm complexity compared to the traditional optimization scheme. At the same time, the introduced binary search algorithm can guide the update of neural network parameters during the training process, which helps the binary search algorithm output a secure coding and power allocation scheme that strictly meets the power constraints and anti-eavesdropping constraints. In order to address the problem that traditional neural networks cannot utilize complex channel parameters, the present invention uses paired real-valued fully connected layers to simulate complex-valued transformations, which can effectively retain the effective information in the complex channel parameters. Secondly, the power constraint is equivalent to the form of an excitation function, which ensures that the secure coding output by the neural network can strictly meet the power constraint. The present invention can significantly improve the system confidentiality rate performance, and can effectively limit the quality of the received signal at the eavesdropper, thereby achieving physical layer secure transmission.

[0148] Based on the above Figure 1 As can be seen from the implementation method, the embodiment of the present invention determines the first signal-to-interference-and-noise ratio (SINR) at the legitimate receiver and the second SINR at the eavesdropper based on the source symbols transmitted on each subcarrier, the security coding matrix in the time domain, the power allocation scheme, the frequency domain channel matrix, and the frequency domain noise vector. Based on the first SINR and the second SINR, a target optimization problem is established, with the signal quality at the legitimate receiver as the optimization target. The target optimization problem is solved using a binary search algorithm and a neural network to obtain the final security coding matrix and the final power allocation scheme. The final security coding matrix and the final power allocation scheme are then used to design a satellite-to-ground network OFDM anti-interception waveform based on security coding. In this way, the target optimization problem established takes the signal quality at the legitimate receiver as the optimization target and utilizes the security coding matrix in the time domain, resulting in higher confidentiality performance for the satellite-to-ground link. The use of a binary search algorithm and a low-complexity neural network can reduce the computing power resource usage of the satellite-to-ground fusion network.

[0149] Based on the same inventive concept, as an implementation of the above-mentioned satellite-to-ground network OFDM anti-interception waveform design method based on security coding, an embodiment of the present invention also provides a satellite-to-ground network OFDM anti-interception waveform design device based on security coding. Figure 9 This is a structural diagram of a satellite-to-ground network OFDM anti-interception waveform design device based on secure coding in an embodiment of the present invention, see Figure 9 As shown, the device may include:

[0150] A determination module 901 is configured to determine a first signal-to-interference-and-noise ratio (SINR) at a legitimate receiver and a second SINR at an eavesdropper based on source symbols transmitted on each subcarrier, a security coding matrix in the time domain, a power allocation scheme, a frequency domain channel matrix, and a frequency domain noise vector;

[0151] An establishing module 902 is configured to establish a target optimization problem based on the first signal to interference plus noise ratio and the second signal to interference plus noise ratio, wherein the target optimization problem takes the signal quality of the legitimate receiver end as an optimization target;

[0152] The solution module 903 is used to solve the target optimization problem using a binary search algorithm and a neural network to obtain a final security coding matrix and a final power allocation scheme, and use the final security coding matrix and the final power allocation scheme to design a satellite-to-ground network OFDM anti-interception waveform based on security coding.

[0153] Establishing module 902, specifically configured to perform an open operation on an average of the first signal to interference plus noise ratio and the second signal to interference plus noise ratio to obtain a symbol error rate;

[0154] Establishing a first optimization problem based on the first signal to interference plus noise ratio and the second signal to interference plus noise ratio, wherein the first optimization problem takes maximizing the confidentiality rate as an optimization goal and the first optimization problem satisfies a symbol error rate constraint;

[0155] The first optimization problem is reconstructed according to the first signal-to-interference-and-noise ratio, the symbol error rate, the transmit power of each subcarrier of the eavesdropper, and the error threshold of the eavesdropper to establish a target optimization problem. The target optimization problem takes the signal quality of the legitimate receiver end as the optimization target.

[0156] Establishing module 902 establishes a first optimization problem based on the first signal to interference plus noise ratio and the second signal to interference plus noise ratio, including: determining a confidentiality rate based on the first signal to interference plus noise ratio and the second signal to interference plus noise ratio; and establishing the first optimization problem based on the confidentiality rate, the transmit power of each subcarrier of the eavesdropper, the error threshold of the eavesdropper, the transmit power of each subcarrier of the legitimate receiver, the error threshold of the legitimate receiver, the transmission power of the subcarrier, the security coding matrix in the frequency domain, and the transmit power upper limit of each subcarrier.

[0157] Establish module 902, the expression of the first optimization problem is:

[0158]

[0159] in, is the first optimization problem, R k is the confidentiality rate of the kth subcarrier, M is the security coding matrix in the frequency domain, p is the power allocation scheme, P b is the transmit power of each subcarrier of the legal receiver, ε b is the error threshold of the legitimate receiver, Pe is the eavesdropper’s transmission power of each subcarrier, ε e is the error threshold of the eavesdropper, p k is the transmission power of the kth subcarrier, n is the nth subcarrier, N is the number of subcarriers, M k,n is the element of the kth row and nth column of the security coding matrix in the frequency domain, P S The upper limit of the transmit power for each subcarrier.

[0160] Establish module 902, the expression of the target optimization problem is:

[0161]

[0162] in, is the target optimization problem, M is the security coding matrix in the frequency domain, p is the power allocation scheme, is the first signal to interference and noise ratio of the kth subcarrier at the legal receiver, P e is the eavesdropper’s transmission power of each subcarrier, ε e is the error threshold of the eavesdropper, p k is the transmission power of the kth subcarrier, n is the nth subcarrier, N is the number of subcarriers, M k,n is the element of the kth row and nth column of the security coding matrix in the frequency domain, P S The upper limit of the transmit power for each subcarrier.

[0163] The solution module 903 is specifically used to decompose the target optimization problem into a security coding optimization problem; input the composite channel state matrix into the complex multi-layer perception mechanism of the neural network to simulate an equivalent complex transformation through the real-valued fully connected layer of the complex multi-layer perception mechanism, and output the optimized security coding matrix in the frequency domain; use the activation function to process the optimized security coding matrix in the frequency domain to obtain the activation function of the optimized security coding matrix in the frequency domain; solve the security coding optimization problem based on the ratio of the optimized security coding matrix in the frequency domain to the activation function of the optimized security coding matrix in the frequency domain, and determine the final security coding matrix; decompose the target optimization problem into a power allocation optimization problem; use the binary search algorithm to solve the power allocation optimization problem to obtain the final power allocation scheme, and use the final security coding matrix and the final power allocation scheme to design a satellite-to-ground network OFDM anti-interception waveform based on security coding.

[0164] Solution module 903 decomposes the target optimization problem into a security coding optimization problem, including: setting the power allocation scheme to a fixed value to decompose the target optimization problem into a security coding optimization problem, where the security coding optimization problem uses the signal quality of the legitimate receiver as the optimization target; decomposing the target optimization problem into a power allocation optimization problem, including: decomposing the target optimization problem into a power allocation optimization problem based on the transmission power of the subcarrier and the minimum subcarrier symbol error rate, where the power allocation optimization problem uses the transmission power of the subcarrier as the objective function.

[0165] In the solution module 903, the expression of the security coding optimization problem is:

[0166]

[0167] in, is a security coding optimization problem, M is a security coding matrix in the frequency domain, is the first signal to interference and noise ratio of the kth subcarrier at the legal receiver, M k,n is the element of the kth row and nth column of the security coding matrix in the frequency domain, P S The transmit power upper limit of each subcarrier, n is the nth subcarrier, and N is the number of subcarriers.

[0168] In the solution module 903, the expression of the power allocation optimization problem is:

[0169]

[0170] in, is the power allocation optimization problem, p k is the transmission power of the kth subcarrier, is the symbol error rate of the first signal to interference and noise ratio of the kth subcarrier at the legal receiver, ε e is the error threshold of the eavesdropper, n is the nth subcarrier, N is the number of subcarriers, M k,n is the element of the kth row and nth column of the security coding matrix in the frequency domain, P S The upper limit of the transmit power for each subcarrier.

[0171] It should be noted that the above description of the embodiment of the device for designing an OFDM waveform for satellite-to-ground network anti-interception based on security coding is similar to the description of the aforementioned method embodiment, and has similar beneficial effects as the embodiment of the method for designing an OFDM waveform for satellite-to-ground network anti-interception based on security coding. For technical details not disclosed in the embodiment of the device for designing an OFDM waveform for satellite-to-ground network anti-interception based on security coding according to the present invention, please refer to the description of the embodiment of the method for designing an OFDM waveform for satellite-to-ground network anti-interception based on security coding according to the present invention.

[0172] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person 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 based on the scope of protection of the claims.

Claims

1. A satellite-to-ground network OFDM anti-interception waveform design method based on secure coding, characterized in that: include: Determine a first signal-to-interference-and-noise ratio (SINR) at a legitimate receiver and a second SINR at an eavesdropper based on source symbols transmitted on each subcarrier, a security coding matrix in the time domain, a power allocation scheme, a frequency domain channel matrix, and a frequency domain noise vector; Establishing a target optimization problem based on the first signal to interference plus noise ratio and the second signal to interference plus noise ratio, wherein the target optimization problem takes signal quality of a legitimate receiver end as an optimization target; The target optimization problem is solved by using a binary search algorithm and a neural network to obtain a final security coding matrix and a final power allocation scheme. The final security coding matrix and the final power allocation scheme are used to design an OFDM anti-interception waveform for a satellite-to-ground network based on security coding.

2. The satellite-to-ground network OFDM anti-interception waveform design method based on secure coding according to claim 1 is characterized in that: The establishing a target optimization problem according to the first signal to interference plus noise ratio and the second signal to interference plus noise ratio includes: performing an open operation on an average of the first signal to interference plus noise ratio and the second signal to interference plus noise ratio to obtain a symbol error rate; Establishing a first optimization problem based on the first signal to interference plus noise ratio and the second signal to interference plus noise ratio, wherein the first optimization problem has maximizing a confidentiality rate as an optimization goal, and the first optimization problem satisfies the constraint of the symbol error rate; The first optimization problem is reconstructed based on the first signal-to-interference-and-noise ratio, the symbol error rate, the transmit power of each subcarrier of the eavesdropper, and the error threshold of the eavesdropper to establish the target optimization problem, where the target optimization problem takes the signal quality of the legitimate receiver end as the optimization target.

3. The satellite-to-ground network OFDM anti-interception waveform design method based on secure coding according to claim 2 is characterized in that: The establishing a first optimization problem according to the first signal to interference plus noise ratio and the second signal to interference plus noise ratio includes: determining the confidentiality rate according to the first signal to interference plus noise ratio and the second signal to interference plus noise ratio; The first optimization problem is established based on the confidentiality rate, the transmission power of each subcarrier of the eavesdropper, the error threshold of the eavesdropper, the transmission power of each subcarrier of the legitimate receiver, the error threshold of the legitimate receiver, the transmission power of the subcarrier, the security coding matrix in the frequency domain and the transmission power upper limit of each subcarrier.

4. The satellite-to-ground network OFDM anti-interception waveform design method based on secure coding according to claim 3 is characterized in that: The expression of the first optimization problem is: in, For the first optimization problem, R k is the confidentiality rate of the kth subcarrier, M is the security coding matrix in the frequency domain, p is the power allocation scheme, P b is the transmit power of each subcarrier of the legal receiver, ε b is the error threshold of the legal receiver, P e is the transmit power of each subcarrier of the eavesdropper, ε e is the error threshold of the eavesdropper, p k is the transmission power of the kth subcarrier, n is the nth subcarrier, N is the number of subcarriers, M k,n is the element of the kth row and nth column of the security coding matrix in the frequency domain, P S is the upper limit of the transmit power of each subcarrier.

5. The satellite-to-ground network OFDM anti-interception waveform design method based on secure coding according to claim 2 is characterized in that: The expression of the target optimization problem is: in, is the target optimization problem, M is the security coding matrix in the frequency domain, p is the power allocation scheme, is the first signal to interference and noise ratio of the kth subcarrier at the legal receiver, P e is the transmit power of each subcarrier of the eavesdropper, ε e is the error threshold of the eavesdropper, p k is the transmission power of the kth subcarrier, n is the nth subcarrier, N is the number of subcarriers, M k,n is the element of the kth row and nth column of the security coding matrix in the frequency domain, P S is the upper limit of the transmit power of each subcarrier.

6. The satellite-to-ground network OFDM anti-interception waveform design method based on secure coding according to claim 5 is characterized in that: The target optimization problem is solved by using a binary search algorithm and a neural network to obtain a final security coding matrix and a final power allocation scheme, and the final security coding matrix and the final power allocation scheme are used to design a satellite-to-ground network OFDM anti-interception waveform based on security coding, including: Decomposing the target optimization problem into a secure coding optimization problem; Inputting the composite channel state matrix into the complex multi-layer perception mechanism of the neural network to simulate an equivalent complex transformation through the real-valued fully connected layer of the complex multi-layer perception mechanism, and outputting an optimized security coding matrix in the frequency domain; Processing the frequency-domain optimized security coding matrix using an activation function to obtain an activation function of the frequency-domain optimized security coding matrix; solving the security coding optimization problem and determining the final security coding matrix according to a ratio of the security coding matrix optimized in the frequency domain to an activation function of the security coding matrix optimized in the frequency domain; Decomposing the target optimization problem into a power allocation optimization problem; The power allocation optimization problem is solved by using the binary search algorithm to obtain the final power allocation scheme, and the satellite-to-ground network OFDM anti-interception waveform based on security coding is designed by using the final security coding matrix and the final power allocation scheme.

7. The satellite-to-ground network OFDM anti-interception waveform design method based on secure coding according to claim 6 is characterized in that: Decomposing the target optimization problem into a secure coding optimization problem includes: Setting the power allocation scheme to a fixed value to decompose the target optimization problem into the security coding optimization problem, wherein the security coding optimization problem takes the signal quality of the legitimate receiver end as an optimization target; Decomposing the target optimization problem into a power allocation optimization problem includes: According to the transmission power of the subcarrier and the minimum symbol error rate of the subcarrier, the target optimization problem is decomposed into the power allocation optimization problem, and the power allocation optimization problem takes the transmission power of the subcarrier as the objective function.

8. The satellite-to-ground network OFDM anti-interception waveform design method based on secure coding according to claim 6 is characterized in that: The expression of the secure coding optimization problem is: in, is the security coding optimization problem, M is the security coding matrix in the frequency domain, is the first signal to interference and noise ratio of the kth subcarrier at the legal receiver, M k,n is the element of the kth row and nth column of the security coding matrix in the frequency domain, P S is the upper limit of the transmission power of each subcarrier, n is the nth subcarrier, and N is the number of subcarriers.

9. The satellite-to-ground network OFDM anti-interception waveform design method based on secure coding according to claim 6, characterized in that: The expression of the power allocation optimization problem is: in, For the power allocation optimization problem, p k is the transmission power of the kth subcarrier, is the symbol error rate of the first signal to interference and noise ratio of the kth subcarrier at the legal receiver, ε e is the error threshold of the eavesdropper, n is the nth subcarrier, N is the number of subcarriers, M k,n is the element of the kth row and nth column of the security coding matrix in the frequency domain, P S is the upper limit of the transmit power of each subcarrier.

10. A satellite-to-ground network OFDM anti-interception waveform design device based on secure coding, characterized in that: include: a determination module, configured to determine a first signal-to-interference-and-noise ratio (SINR) at a legitimate receiver and a second SINR at an eavesdropper based on source symbols transmitted on each subcarrier, a security coding matrix in the time domain, a power allocation scheme, a frequency domain channel matrix, and a frequency domain noise vector; an establishing module, configured to establish a target optimization problem based on the first signal to interference plus noise ratio and the second signal to interference plus noise ratio, wherein the target optimization problem takes the signal quality of the legitimate receiver end as an optimization target; A solution module is used to solve the target optimization problem using a binary search algorithm and a neural network to obtain a final security coding matrix and a final power allocation scheme, and use the final security coding matrix and the final power allocation scheme to design a satellite-to-ground network OFDM anti-interception waveform based on security coding.

Images