Secure communication method for RIS unknown eavesdropping channel

Abstract

The invention relates to a secure communication method for an RIS unknown eavesdropping channel, and belongs to the technical field of secure communication. The method comprises the following steps: determining equivalent channel parameters of a direct channel of a base station-legal receiver and a base station-RIS-legal receiver reflection channel by using pilot frequency information fed back by the legal receiver, and constructing an RIS secure communication model assisted by artificial noise based on a preset pilot frequency codebook; determining the channel capacity of a legal receiver, and determining the probability distribution information of an eavesdropping signal-to-noise ratio based on the statistical characteristics of an eavesdropping channel; and constructing a joint optimization problem of an information bearing signal transmitted by a base station, an artificial noise signal and an RIS phase shift matrix by taking the obtained confidential outage probability minimization of the RIS communication system as an optimization target, and decomposing the joint optimization problem into two layers of optimization problems to solve so as to realize the secure communication of the RIS system. The invention aims to solve the technical problem that the security of the RIS communication system is difficult to ensure when the channel degree of freedom is limited and the eavesdropping channel state information is unknown in the prior art.

Description

Technical Field

[0001] This invention relates to a secure communication method for unknown eavesdropping channels of RIS, belonging to the field of secure communication technology. Background Technology

[0002] Artificial noise technology has become an important means of ensuring the security of wireless communication systems and is one of the key strategies for preventing eavesdropping. Artificial noise technology introduces additional noise signals at the transmitting end, utilizing the spatial degrees of freedom of the legitimate channel state information matrix to ensure that the noise vector remains orthogonal to the legitimate channel state information matrix. Due to the channel difference between the eavesdropping channel matrix and the legitimate channel matrix, legitimate receivers can correctly decode the information-carrying signal, while eavesdroppers find it difficult to extract useful information from the composite signal, effectively reducing the channel capacity of the eavesdropping channel. However, the core requirement for implementing artificial noise technology is that the number of transmitting antennas at the base station must be greater than the number of receiving antennas at the legitimate receiver. This condition is necessary because the number of transmitting antennas determines the spatial degrees of freedom of the wireless communication system, and the higher the spatial degrees of freedom, the larger the dimension of noise signals that can be accommodated. With the widespread application of new wireless communication systems such as reconfigurable smart surfaces, the antenna configuration and link structure of wireless communication systems have become more complex. When the number of transmitting antennas is less than or equal to the number of receiving antennas, it is difficult to find a suitable null space to accommodate the artificial noise vector. Furthermore, reconfigurable smart surface technology typically involves the coordinated operation of multiple links, requiring simultaneous consideration of noise design requirements in both the uplink and downlink. In this multi-link scenario, the artificial noise signal needs to maintain orthogonality with both the uplink and downlink, which poses a greater challenge to the design of the artificial noise vector.

[0003] As a non-cooperative party in a reconfigurable smart surface communication system, the eavesdropper's eavesdropping channel state information is often difficult for legitimate communicating parties to accurately obtain, which limits the applicability and effectiveness of artificial noise theoretical models. Furthermore, in a reconfigurable smart surface communication system, the eavesdropping link consists of two unknown channels, further increasing the complexity of artificial noise model analysis. Considering the random characteristics of wireless channels, wireless channel fading is one of the key factors affecting signal transmission quality. Introducing a random channel fading model can more accurately reflect the random fluctuations and attenuation effects of the eavesdropping channel, providing strong support for the performance analysis and theoretical characterization of artificial noise schemes. Based on the time-varying characteristics of legitimate channels, legitimate communicating parties can construct pilot codebooks and use the codebook index to transmit noise signals, employing a "one frame, one key" scheme to effectively overcome the problem of artificial noise being limited by the channel's degrees of freedom. Summary of the Invention

[0004] The purpose of this invention is to provide a secure communication method for RIS with unknown eavesdropping channels, aiming to solve the technical problem that it is difficult to guarantee the security of RIS communication systems when the channel freedom is limited and the eavesdropping channel status information is unknown.

[0005] To achieve the above objectives, the technical solution of this invention is: a secure communication method for unknown eavesdropping channels of RIS (Reference Signals and Information). This method constructs an artificial noise-assisted reconfigurable smart surface secure communication model based on a preset pilot codebook. Utilizing the time-varying characteristics of the channel, legitimate communicating parties construct a pilot codebook, transmit noise signals through codebook indexing, and employs gamma distribution to statistically model the eavesdropping channel to minimize the probability of security interruption and evaluate the system's security performance. The method includes the following steps: Step 1: Using the pilot information fed back by the legitimate receiver, determine the equivalent channel parameters of the direct channel between the base station and the legitimate receiver, as well as the reflection channel between the base station, the RIS, and the legitimate receiver, and construct an artificial noise-assisted RIS secure communication model based on the preset pilot codebook. Step 2: Based on the RIS secure communication model, determine the channel capacity of the legitimate receiver, and under the condition that the instantaneous channel state information of the eavesdropper is unknown, determine the probability distribution information of the eavesdropping signal-to-noise ratio based on the statistical characteristics of the eavesdropping channel; Step 3: Based on the channel capacity of the legitimate receiver, the probability distribution information of the eavesdropping signal-to-noise ratio, and the preset confidentiality rate threshold, the confidentiality interruption probability of the RIS communication system is obtained; Step 4: With minimizing the probability of secure interruption in the RIS communication system as the optimization objective, under the constraints of the total transmit power of the base station and the unit modulus of the RIS reflection unit, a joint optimization problem is constructed for the information-bearing signal transmitted by the base station, the artificial noise signal, and the RIS phase shift matrix. The joint optimization problem is then decomposed into two-level optimization problems to achieve secure communication of the RIS system. The first-level optimization problem is to solve for the power distribution coefficient between the artificial noise signal and the information-carrying signal under the condition of a fixed RIS phase shift matrix. The second-level optimization problem is to solve the RIS phase shift matrix under the condition of fixed power allocation coefficient.

[0006] Optionally, Step 1 specifically includes: First, the RIS secure communication model is defined as consisting of a base station, RIS, target receiver, and eavesdropper; where the target receiver of the information sent by the base station in the current time slot is defined as the legitimate receiver, and other receivers are defined as eavesdroppers. Secondly, the legitimate receiver will combine the channel state information matrix. Feedback is sent back to the base station, and calculations are performed. Frobenius norm The legitimate recipient is based on The value is used to determine the artificial noise signal transmitted by the base station in the next time slot from the preset pilot codebook; wherein, the expression of the composite channel state information matrix is:

[0007] In the formula, , and Let represent the channel state information matrix between the base station and the legitimate receiver, the channel state information matrix between the base station and the RIS, and the channel state information matrix between the RIS and the legitimate receiver, respectively. It is the channel state information matrix The probability of its existence. Represents the phase shift matrix of RIS; Then, the base station receives the feedback composite channel state information matrix, calculates the Frobenius norm value, and determines the artificial noise signal to be transmitted in the next time slot based on the preset pilot codebook. Finally, in the next time slot, the legitimate receiver receives a composite signal containing both artificial noise and the information-carrying signal, according to... The value selects an artificial noise decoding method from the preset pilot codebook to cancel the interference of artificial noise signals on the signals carried by the decoded information of legitimate receivers.

[0008] Optionally, the channel capacity of the legitimate receiver is specifically:

[0009] in, The channel capacity represents the capacity of legitimate receivers. Represents the information-carrying signal vector. This represents the variance of the signal carrying the information. This represents the variance of additive white Gaussian noise in a legitimate channel.

[0010] Optionally, the probability distribution information of the eavesdropping signal-to-noise ratio specifically includes:

[0011] in, , , Let be the probability density function. Indicates the signal-to-noise ratio (SNR) in eavesdropping. and These represent the channel state information matrix between the base station and the eavesdropper, and the channel state information matrix between the RIS and the eavesdropper, respectively. yes The probability of its existence. Let be an artificial noise signal vector, satisfying , It is the variance of the artificial noise signal. , The number of eavesdropping antennas, The parameter is beta function, Let be the scale parameter, where and .

[0012] Optionally, the specific method for obtaining the confidentiality interruption probability of the RIS communication system is as follows:

[0013] in, This represents the probability of a secure interruption in a RIS communication system. To eavesdrop on the signal-to-noise ratio threshold, This indicates the physical layer coding rate.

[0014] Optionally, the joint optimization problem of constructing the information-bearing signal, artificial noise signal, and RIS phase shift matrix transmitted by the base station is specifically as follows:

[0015] in, The phase shift matrix of RIS The Middle OK The value of the column, Indicates the first The phase angle of each reflecting unit, This indicates the total power transmitted by the base station. This represents the number of reflection units in the RIS.

[0016] Optionally, the first-level optimization problem is specifically:

[0017] In the formula, This is the maximum score. right Differentiate to obtain the power distribution coefficient. ,in, , .

[0018] Optionally, the second-level optimization problem is specifically as follows:

[0019] in, , For constant terms, Represents the trace of a matrix. Represents the conjugate transpose of a matrix. It is a positive semi-definite matrix. , , , The phase shift matrix of RIS A vector of diagonal elements , Therefore A diagonal matrix with diagonal elements. To represent the transpose of a matrix, , , This represents the vector obtained by taking the complex conjugate of each element. , , This represents the vector obtained by transposing each element. Indicates taking OK The elements of the columns form a matrix. Denotes the rank of a matrix and the relaxation constraints. .

[0020] The beneficial effects of this invention are as follows: Legitimate communicating parties construct a pilot codebook using shared pilot information. The base station determines the noise signal transmitted in each time slot based on the pilot codebook. Upon receiving the composite signal, the legitimate receiver queries the same codebook to obtain the decoding method for the noise components, reducing interference to the legitimate receiver, improving the adaptability of artificial noise technology to changes in the degrees of freedom of the legitimate channel, and enhancing system confidentiality. Furthermore, this invention utilizes gamma distribution to statistically model the eavesdropping channel, effectively solving the problem that the base station cannot effectively optimize power allocation when the eavesdropping channel state information is unknown. It further analyzes and proves that the probability density function of the eavesdropping signal-to-noise ratio follows a beta prime distribution, and the asymptotic expression for minimizing the probability of confidentiality interruption presents a regularized incomplete beta function. Finally, it integrates alternating optimization algorithms and semi-definite relaxation algorithms to solve for the optimal power allocation and local optimal solutions of the phase shift matrix of the system. Attached Figure Description

[0021] Figure 1 This is a system model diagram of the present invention; Figure 2 This is a flowchart of the algorithm of the present invention; Figure 3 This is a graph showing the trend of the upper limit of integration under different signal-to-noise ratios and physical layer coding rates of the present invention; Figure 4 This is a graph showing the trend of the probability of security interruption under different signal-to-noise ratios and physical layer coding rates. Figure 5 This is a convergence analysis diagram of the algorithm of this invention. Detailed Implementation

[0022] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0023] Example 1: A secure communication method for unknown eavesdropping channels of RIS, comprising the following steps: Step 1: Using the pilot information fed back by the legitimate receiver, determine the equivalent channel parameters of the direct channel between the base station and the legitimate receiver, as well as the reflection channel between the base station, the RIS, and the legitimate receiver, and construct an artificial noise-assisted RIS secure communication model based on the preset pilot codebook. Optionally, the base station and the legitimate receiver jointly calculate and statistically analyze the Frobenius norm of the legitimate channel state information using pilot information, and construct a pilot codebook containing the encoding and decoding methods of artificial noise signals that is known only to the two legitimate communicating parties. The base station obtains the artificial noise signal to be transmitted in the next time slot by querying the pilot codebook. After receiving the composite signal containing the information-bearing signal and the artificial noise signal, the legitimate receiver also obtains the decoding method of the artificial noise signal by querying the pilot codebook, thus realizing "one frame, one key" information transmission.

[0024] Optionally, such as Figure 1 As shown, the Reconfigurable Smart Surface (RIS) system consists of a base station (BS), a legitimate receiver (Bob), and an eavesdropper (Eve). The BS is equipped with... Bob is equipped with a single transmitting antenna. Eve is equipped with a root receiving antenna. Root eavesdropping antenna, RIS by It consists of several controllable reflection units. The channel state information (CSI) of the BS-RIS link is denoted as... The CSI values ​​for the RIS-Bob and RIS-Eve links are respectively... and The CSI values ​​for BS-Bob and BS-Eve respectively represent... and . , and Both are bidirectional channels used to feed back the CSI obtained from the pilot signals. Due to obstruction, and It doesn't always exist. Definition and They represent and The probability of its existence. The phase shift coefficient matrix of RIS is defined as... ,in, This indicates extracting the diagonal elements of the matrix. Indicates the first The phase angle of each reflecting unit.

[0025] Furthermore, since Eve is a non-cooperative node, its CSI is unknown. In the absence of prior knowledge, based on the principle of maximum entropy, we will... and Element-wise modeled as independent and identically distributed circular symmetric complex Gaussian channels, satisfying: and ,in, Represents the vectorization of a matrix. This indicates that it follows a complex Gaussian distribution. Represents the identity matrix.

[0026] To prevent Eve from obtaining information transmitted through legitimate channels, in the first... In each time slot, BS transmits information bearer signals to Bob. At the same time, send a message with irrelevant artificial noise signals Step 1 specifically includes: First, the RIS secure communication model is defined as consisting of a base station, RIS, target receiver, and eavesdropper; where the target receiver of the information sent by the base station in the current time slot is defined as the legitimate receiver, and other receivers are defined as eavesdroppers. Secondly, the legitimate receiver will combine the channel state information matrix. Feedback is sent back to the base station, and calculations are performed. Frobenius norm The legitimate recipient is based on The value is used to determine the artificial noise signal transmitted by the base station in the next time slot from the preset pilot codebook; wherein, the expression of the composite channel state information matrix is: ; Then, the base station receives the feedback composite channel state information matrix, calculates the Frobenius norm value, and determines the artificial noise signal to be transmitted in the next time slot based on the preset pilot codebook. Finally, in the next time slot, the legitimate receiver receives a composite signal containing both artificial noise and the information-carrying signal, according to... The value selects an artificial noise decoding method from the preset pilot codebook to cancel the interference of artificial noise signals on the signals carried by the decoded information of legitimate receivers.

[0027] Optionally, in this embodiment, the encoding method of the pilot codebook can be selected from a series of mature channel coding schemes. The encoding techniques that can be used include, but are not limited to, polar codes, low-density parity-check codes, Turbo codes and their variants, to adapt to the complexity and performance requirements of different wireless communication systems.

[0028] Step 2: Based on the RIS secure communication model, determine the channel capacity of the legitimate receiver, and under the condition that the instantaneous channel state information of the eavesdropper is unknown, determine the probability distribution information of the eavesdropping signal-to-noise ratio based on the statistical characteristics of the eavesdropping channel; Optionally, legitimate communicating parties can effectively eliminate [the virus] at the legitimate receiving end by using the pilot codebook. However, due to channel differences, eavesdroppers cannot obtain codebook information to decode the code. The signals received by Bob and Eve are as follows: (1) (2) In the formula, and Let the channel state information matrix of the base station-eavesdropper and the channel state information matrix of the RIS-eavesdropper be respectively, and let... Information-carrying symbols Precoding vector ,satisfy . , Let be an artificial noise signal vector, satisfying , . and Let the additive white Gaussian noise of the legitimate link and the eavesdropping link be respectively, satisfying the following conditions: and , Let V be the variance of additive white Gaussian noise in a legitimate channel. Then the system security rate is: (3) In the formula, and These are the channel capacities for legitimate communication links and eavesdropping communication links, respectively. This represents the variance of the signal carrying the information. It is the variance of the artificial noise signal.

[0029] It's important to understand that this is because BS and Bob cannot obtain... and Precise information is therefore difficult to obtain through optimization. , and The way makes Maximum. Therefore, this embodiment introduces a physical layer coding rate. The probability of a security breach is used as a measure of the security performance of noise signals. (4) In the formula, Represents the probability of an event occurring, when As the value increases, the system's anti-interference capability decreases and the bit error rate increases. It represents the probability of secure communication of the system under the premise of reliable transmission through a legitimate channel.

[0030] Therefore, the secure communication problem in this embodiment can be formally described as: through joint optimization , and To minimize the probability of security breach, the expression is: (5) in, The phase shift matrix of RIS The Middle OK The value of the column, This indicates the total power transmitted by the base station. This represents the number of reflection units in the RIS.

[0031] Step 3: Based on the channel capacity of the legitimate receiver, the probability distribution information of the eavesdropping signal-to-noise ratio, and the preset confidentiality rate threshold, the confidentiality interruption probability of the RIS communication system is obtained; Optionally, the eavesdropping signal-to-noise ratio is defined as: (6) in, , The parameter is beta function, This is the scale parameter.

[0032] Furthermore, set for 3D complex vector and Dimensions and Independent and identically distributed complex Gaussian random variables, Let represent a constant taking values ​​from 0 to 1. Then the random variable... The mean and variance are: (7) (8) Based on the parametric properties of the gamma distribution, the shape parameter and scale parameter of the gamma distribution are as follows: (9) (10) According to the definition of the gamma distribution, random variables satisfy ,but The probability density function is: (11) In the formula, Represents the gamma function. For shape parameters, Let be the scale parameter. According to formula (6), let . , Then the random variable and The probability density functions are as follows: (12) (13) In the formula, , When the eavesdropping link is under high signal-to-noise ratio conditions ( When eavesdropping, the signal-to-noise ratio is... The probability density function is: (14) In the formula, according to the definition of the gamma function .make , ,get: (15) In the formula, According to the definition of beta prime distribution: (16) when At that time, the signal-to-noise ratio of eavesdropping Obeying shape parameters Scale parameters are The distribution of beta primes. (From...) The probability of system security breach is expressed as: (17) make ,get , and The asymptotic expression for minimizing the security breach probability of a reconfigurable smart surface system can be expressed as: (18) Based on the additivity of the definite integral interval, we derive that the asymptotic form is a regularized incomplete beta function: (19) In the formula, , This is the threshold value for the signal-to-noise ratio used in eavesdropping. This represents the incomplete regularization beta function.

[0033] Step 4: With minimizing the probability of secure interruption in the RIS communication system as the optimization objective, under the constraints of the total transmit power of the base station and the unit modulus of the RIS reflection unit, a joint optimization problem is constructed for the information-bearing signal transmitted by the base station, the artificial noise signal, and the RIS phase shift matrix. The joint optimization problem is then decomposed into two-level optimization problems to achieve secure communication of the RIS system. The first-level optimization problem is: under the condition of a fixed RIS phase shift matrix, to solve for the power allocation coefficient between the artificial noise signal and the information-carrying signal, specifically: Alternatively, as shown in formula (19), minimizing the probability of security breach is equivalent to maximizing the regularized incomplete beta function. Therefore, this embodiment demonstrates... Regarding the points cap The monotonicity of the beta function determines the boundary of the feasible region where the optimal solution is located. Define the regularized incomplete beta function: (20) In the formula, In the interval The partial derivative within can be expressed as: (twenty one) In the formula, In the interval The inner monotonically increasing condition. Therefore, the nonconvex problem of minimizing the probability of security breach in the system is equivalent to maximizing the upper bound of the integral. The problem of finding the maximum and minimum values ​​is represented by the optimization problem as follows: (twenty two) According to formula (3) Substituting the expression into formula (19), we obtain the upper limit of integration. for: (twenty three) In the formula, .

[0034] Furthermore, when When fixed, let , .in For information carrying signals and noise signals The power allocation factor between them. Definition , and ,but Represented as: (twenty four) The first derivative is: (25) make ,but Due to constraints It can be seen that if and only if At that time, the following conditions are met: , . At point Take the maximum value at that point.

[0035] The second-level optimization problem is to solve the RIS phase shift matrix under a fixed power allocation coefficient, specifically: Substitute the optimal power allocation coefficient, according to The quadratic form of a vector is defined as follows: , ,therefore Let be a constant. , objective function Represented as: (26) In the formula, let ,but: (27) From the relationship between vector norm and matrix trace: .make , ,but Expanded to: (28) In the formula , . For constant terms, Represents the trace of a matrix. Represents the conjugate transpose of a matrix. This represents the transpose of a matrix. Using the linear property of matrices, it satisfies: , and . for A vector of diagonal elements. This represents the vector obtained by taking the complex conjugate of each element. This represents the vector obtained by transposing each element. for A vector of diagonal elements Therefore The diagonal elements are a diagonal matrix. Therefore, the objective function can be rewritten as: (29) make , , , Its objective function is: The optimization problem then becomes: (30) In the formula, It is a positive semi-definite matrix. Indicates taking OK The elements of the columns form a matrix. Denotes the rank of a matrix and the relaxation constraints. Finally, the solution was obtained using MATLAB's CVX toolkit.

[0036] Understandably, this embodiment constructs an artificial noise-assisted RIS secure communication model based on a preset pilot codebook in Step 1, thus solving the problem of artificial noise signals being affected by channel degrees of freedom. Secondly, Step 2 determines the legitimate channel capacity, which is easily obtained for legitimate receivers. However, eavesdroppers are non-cooperative, and neither the base station nor the legitimate receiver is certain of the eavesdropper's instantaneous channel state information. Therefore, a channel model based on statistical eavesdropping is constructed. Then, the probability distribution information of the eavesdropping signal-to-noise ratio is derived, and a statistical eavesdropping channel model is constructed using gamma distribution, as gamma distribution can effectively represent the sum of multiple independent exponentially distributed random variables, making it suitable for RIS channel modeling. Then, Step 3 determines the expression for the probability of security interruption. Finally, Step 4 constructs an optimization model based on the objective function to achieve secure communication in the RIS system.

[0037] The invention will now be further described with reference to the accompanying drawings.

[0038] Figure 2 This is the overall algorithm flowchart for this invention. First, input initial values: including the number of transmitting antennas. Number of receiving antennas and the number of controllable units And set the initial number of iterations. Maximum number of iterations Phase shift matrix Power allocation coefficient and convergence tolerance Then an iterative loop is performed. Through formula Calculate the optimal power allocation point and substitute the updated result into the objective function. Use a positive semidefinite relaxation algorithm to solve for the phase shift matrix until the convergence criterion is met. Final output result , and .

[0039] Figure 3 The probability of existence of the present invention in different direct-light channels and physical layer coding rate Points increase The trend of changes in the legal signal-to-noise ratio. (From...) It can be seen that, The smaller, The higher the probability of it being obscured or unusable, The downward trend is more pronounced; however, as the signal-to-noise ratio increases, This indicates that even No, under good legal channel conditions, the use of pilot codebook-based artificial noise technology can still guarantee secure communication for reconfigurable smart surface systems. Furthermore, as the physical layer coding rate increases, the eavesdropping signal-to-noise ratio threshold... It can be seen that, reduce, Reducing noise places higher demands on the robustness and parameter configuration of artificial noise, thus requiring a smaller physical layer coding rate to ensure that the receiver has sufficient noise immunity and error correction capability.

[0040] Figure 4 To address the issue of the number of eavesdropping antennas in this invention The curves showing the variation of the minimum security interruption probability under different legal signal-to-noise ratios. This varies with the physical layer coding rate. As the channel condition decreases, the probability of security breach gradually decreases. Furthermore, the better the channel conditions, the more significant the trend. According to... This verifies that minimizing the probability of security breach is equivalent to maximizing the integral upper limit. In addition, combined with Figure 3 It can be seen that, with the physical layer coding rate The decrease in the upper limit of integration As the system's security is increased, the probability of security breach will gradually approach 0.

[0041] Figure 5 This invention verifies the convergence of the alternating optimization algorithm and the semidefinite relaxation algorithm in reconfigurable smart surface systems. Simulation results show that the upper limit of integration increases with the number of iterations. Gradually increase, power distribution coefficient The gradual decrease and maintenance of a value greater than 0.1 indicates that the system needs to allocate more power to the noisy signal to ensure the confidentiality of the wireless communication system. Furthermore, the curve demonstrates good convergence, with the convergence point occurring before the 20th iteration, verifying the effectiveness of the algorithm of this invention.

[0042] The specific embodiments of the present invention described above in conjunction with the accompanying drawings are only used to illustrate the technical principles and implementation methods of the present invention, and do not constitute a limitation on the scope of protection of the present invention. For those skilled in the art, various equivalent modifications or substitutions can be made to the embodiments based on the concept of the present invention without departing from the spirit and essence of the present invention, and all such modifications or substitutions should be covered within the scope of protection of the present invention.

Claims

1. A secure communication method for unknown eavesdropping channels of RIS, characterized in that, The method includes the following steps: Step 1: Using the pilot information fed back by the legitimate receiver, determine the equivalent channel parameters of the direct channel between the base station and the legitimate receiver, as well as the reflection channel between the base station, the RIS, and the legitimate receiver, and construct an artificial noise-assisted RIS secure communication model based on the preset pilot codebook. Step 2: Based on the RIS secure communication model, determine the channel capacity of the legitimate receiver, and under the condition that the instantaneous channel state information of the eavesdropper is unknown, determine the probability distribution information of the eavesdropping signal-to-noise ratio based on the statistical characteristics of the eavesdropping channel; Step 3: Based on the channel capacity of the legitimate receiver, the probability distribution information of the eavesdropping signal-to-noise ratio, and the preset confidentiality rate threshold, the confidentiality interruption probability of the RIS communication system is obtained; Step 4: With minimizing the probability of secure interruption in the RIS communication system as the optimization objective, under the constraints of the total transmit power of the base station and the unit modulus of the RIS reflection unit, a joint optimization problem is constructed for the information-bearing signal transmitted by the base station, the artificial noise signal, and the RIS phase shift matrix. The joint optimization problem is then decomposed into two-level optimization problems to achieve secure communication of the RIS system. The first-level optimization problem is to solve for the power distribution coefficient between the artificial noise signal and the information-carrying signal under the condition of a fixed RIS phase shift matrix. The second-level optimization problem is to solve the RIS phase shift matrix under the condition of fixed power allocation coefficient.

2. The secure communication method for an unknown eavesdropping channel of RIS according to claim 1, characterized in that, Step 1 specifically refers to: First, the RIS secure communication model is defined as consisting of a base station, RIS, target receiver, and eavesdropper; where the target receiver of the information sent by the base station in the current time slot is defined as the legitimate receiver, and other receivers are defined as eavesdroppers. Secondly, the legitimate receiver will combine the channel state information matrix. Feedback is sent back to the base station, and calculations are performed. Frobenius norm The legitimate recipient is based on The value is used to determine the artificial noise signal transmitted by the base station in the next time slot from the preset pilot codebook; wherein, the expression of the composite channel state information matrix is: ； In the formula, , and These represent the channel state information matrix between the base station and the legitimate receiver, the channel state information matrix between the base station and the RIS, and the channel state information matrix between the RIS and the legitimate receiver, respectively. It is the channel state information matrix The probability of its existence. Represents the phase shift matrix of RIS; Then, the base station receives the feedback composite channel state information matrix, calculates the Frobenius norm value, and determines the artificial noise signal to be transmitted in the next time slot based on the preset pilot codebook. Finally, in the next time slot, the legitimate receiver receives a composite signal containing both artificial noise and the information-carrying signal, according to... The value selects an artificial noise decoding method from the preset pilot codebook to cancel the interference of artificial noise signals on the signals carried by the decoded information of legitimate receivers.

3. A secure communication method for an unknown eavesdropping channel of RIS according to claim 2, characterized in that, The channel capacity of the legitimate receiver is specifically: ； in, The channel capacity represents the capacity of legitimate receivers. Represents the information-carrying signal vector. This represents the variance of the signal carrying the information. This represents the variance of additive white Gaussian noise in a legitimate channel.

4. A secure communication method for an unknown eavesdropping channel of RIS according to claim 3, characterized in that, The probability distribution information of the eavesdropping signal-to-noise ratio is specifically as follows: ； in, , , Let be the probability density function. Indicates the signal-to-noise ratio (SNR) in eavesdropping. and These represent the channel state information matrix between the base station and the eavesdropper, and the channel state information matrix between the RIS and the eavesdropper, respectively. yes The probability of its existence. Let be an artificial noise signal vector, satisfying , It is the variance of the artificial noise signal. , The number of eavesdropping antennas, The parameter is beta function, Let be the scale parameter, where and .

5. A secure communication method for an unknown eavesdropping channel of RIS according to claim 4, characterized in that, The specific probability of secure interruption in the RIS communication system is as follows: ； in, This represents the probability of a secure interruption in a RIS communication system. To eavesdrop on the signal-to-noise ratio threshold, This indicates the physical layer coding rate.

6. A secure communication method for an unknown eavesdropping channel of RIS according to claim 5, characterized in that, The joint optimization problem of constructing the information-bearing signal, artificial noise signal, and RIS phase shift matrix transmitted by the base station is specifically as follows: ； in, The phase shift matrix of RIS The Middle OK The value of the column, Indicates the first The phase angle of each reflecting unit, This indicates the total power transmitted by the base station. This represents the number of reflection units in the RIS.

7. A secure communication method for an unknown eavesdropping channel of RIS according to claim 6, characterized in that, The first-level optimization problem is specifically: ； In the formula, This is the maximum score. right Differentiate to obtain the power distribution coefficient. ,in, , .

8. A secure communication method for an unknown eavesdropping channel of RIS according to claim 7, characterized in that, The second-level optimization problem is specifically as follows: ； in, , For constant terms, Represents the trace of a matrix. This represents the conjugate transpose of a matrix. It is a positive semi-definite matrix. , , , The phase shift matrix of RIS A vector of diagonal elements , Therefore A diagonal matrix with diagonal elements. Represents the transpose of a matrix. , , This represents the vector obtained by taking the complex conjugate of each element. , , This represents the vector obtained by transposing each element. Indicates taking OK The elements of the columns form a matrix. Describing the rank of a matrix and relaxing constraints. .

Images