Intelligent anti-collusion eavesdropping secure communication method, device, equipment and medium

Abstract

The application provides a kind of intelligent anti-collaborative eavesdropping secure communication method, device, equipment and medium. It relates to the field of communication technology. The method is applied to the transmission system composed of base station and reconfigurable intelligent surface, comprising: obtaining the first state information, the first beamforming matrix and the first phase matrix of a plurality of communication links at the current time step;Based on the first state information, the first beamforming matrix and the first phase matrix, the joint action value at the current time step is obtained;Based on the joint action value, the first beamforming matrix and the first phase matrix, with the optimization target of maximizing the private transmission rate, the second beamforming matrix and the second phase matrix at the next time step are obtained;Based on the second beamforming matrix, the base station is adjusted, and based on the second phase matrix, the reconfigurable intelligent surface is adjusted. Through the application, the effect of fast and effective anti-eavesdropping is achieved.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to an intelligent anti-cooperative eavesdropping security communication method, device, equipment and medium. Background Technology

[0002] Because wireless communication signals have broadcast characteristics, information transmitted wirelessly is vulnerable to eavesdropping and illegal interception during transmission. With the increasing number of devices using wireless communication and the growing density of wireless communication network deployments, there are issues with the effective protection of information privacy and security.

[0003] Among related technologies, the method of preventing eavesdropping by injecting artificial noise (AN) involves introducing interference signals into the base station's transmitted signals to disrupt the eavesdropping link, but this method has the problem of poor anti-eavesdropping effectiveness.

[0004] Therefore, there is an urgent need for a secure and effective transmission solution based on reconfigurable smart surfaces to achieve fast and effective anti-eavesdropping. Summary of the Invention

[0005] This application provides a smart anti-collusive eavesdropping security communication method, device, equipment, and medium to achieve a fast and effective anti-eavesdropping effect.

[0006] In a first aspect, this application provides an intelligent anti-cooperative eavesdropping security communication method, applied to a transmission system composed of a base station and a reconfigurable smart surface, comprising:

[0007] Acquire the first state information, first beamforming matrix, and first phase matrix of multiple communication links at the current time step. The communication links include the communication links of legitimate users and the communication links of eavesdroppers.

[0008] Based on the first state information, the first beamforming matrix, and the first phase matrix, the joint action value at the current time step is obtained;

[0009] Based on the joint action value, the first beamforming matrix, and the first phase matrix, with the optimization objective of maximizing the privacy transmission rate, the second beamforming matrix and the second phase matrix for the next time step are obtained.

[0010] The base station is adjusted based on the second beamforming matrix, and the reconfigurable smart surface is adjusted based on the second phase matrix.

[0011] In one possible implementation, based on the first state information, the first beamforming matrix, and the first phase matrix, and with maximizing the privacy transmission rate as the optimization objective, the joint action value at the current time step is obtained, including:

[0012] Based on the first state information, determine the achievable confidentiality rate at the current time step and obtain the immediate reward corresponding to the current time step;

[0013] The immediate reward, the first beamforming matrix, and the first phase matrix corresponding to the current time step are used as inputs to the joint action determination model to obtain the joint action value of the current time step.

[0014] In one possible implementation, after adjusting the base station based on the second beamforming matrix and adjusting the reconfigurable smart surface based on the second phase matrix, the method further includes:

[0015] Based on the second beamforming matrix and the second phase matrix, the second state information of multiple communication links in the next time step is obtained;

[0016] By integrating the first state information, the second state information, the first beamforming matrix, the first phase matrix, the second beamforming matrix, and the second phase matrix, empirical data for the current time step is obtained.

[0017] Experience data is stored in an experience pool.

[0018] In one possible implementation, the immediate reward, the first beamforming matrix, and the first phase matrix corresponding to the current time step are used as inputs to the joint action determination model to obtain the joint action value of the current time step:

[0019] Based on immediate rewards, the joint actions determine the objective function of the model at the current time step;

[0020] Based on the objective function, the first beamforming matrix, and the first phase matrix, the loss function of the joint action determination model at the current time step is determined.

[0021] Based on the objective function and loss function, the joint action determination model is trained using the gradient descent algorithm to obtain the joint action value at the current time step.

[0022] In one possible implementation, based on the joint action value, the first beamforming matrix, and the first phase matrix, the second beamforming matrix and the second phase matrix for the next time step are obtained, including:

[0023] Based on the joint action value of the current time step, determine the beamforming network gradient and phase control network gradient corresponding to the current time step;

[0024] Based on the gradients of multiple communication links and beamforming network at the current time step, the second beamforming matrix for the next time step is obtained through the beamforming network.

[0025] Based on the gradients of multiple communication links and the phase control network at the current time step, the second phase matrix for the next time step is obtained through the phase control network.

[0026] In one possible implementation, based on multiple communication links at the current time step, the achievable security rate at the current time step is determined, and the instantaneous reward corresponding to the current time step is obtained, including:

[0027] Based on the product of large-scale fading and small-scale fading of each communication link, an equivalent baseband channel is constructed for each communication link, resulting in multiple equivalent baseband channels. Among them, the equivalent baseband channels include direct equivalent baseband channels, indirect equivalent baseband channels, reconstructed equivalent baseband channels, direct eavesdropping equivalent baseband channels, and reconstructed eavesdropping equivalent baseband channels.

[0028] Based on the direct equivalent baseband channel, the indirect equivalent baseband channel, the reconstructed equivalent baseband channel, and the additive complex Gaussian noise corresponding to each user, the achievable transmission rate for each user is obtained.

[0029] Based on the indirect equivalent baseband channel, the direct eavesdropping equivalent baseband channel, the reconstructed eavesdropping equivalent baseband channel, and the additive complex Gaussian noise corresponding to the eavesdropper, the achievable eavesdropping rate corresponding to each eavesdropper is obtained.

[0030] Based on the achievable transmission rate and achievable eavesdropping rate, determine the achievable security rate for the current time step and obtain the instant reward corresponding to the current time step.

[0031] Secondly, this application provides a transmission device based on a reconfigurable smart surface, deployed in a transmission system composed of a base station and a reconfigurable smart surface, comprising:

[0032] The acquisition module is used to acquire the first state information, the first beamforming matrix, and the first phase matrix of multiple communication links at the current time step. The communication links include the communication links of legitimate users and the communication links of eavesdroppers.

[0033] The processing module is used to obtain the joint action value of the current time step based on the first state information, the first beamforming matrix, and the first phase matrix; based on the joint action value, the first beamforming matrix, and the first phase matrix, and with the optimization objective of maximizing the privacy transmission rate, obtain the second beamforming matrix and the second phase matrix of the next time step; adjust the base station based on the second beamforming matrix, and adjust the reconfigurable smart surface based on the second phase matrix.

[0034] In one possible implementation, the processing module is specifically used for:

[0035] Based on the first state information, determine the achievable confidentiality rate at the current time step and obtain the immediate reward corresponding to the current time step;

[0036] The immediate reward, the first beamforming matrix, and the first phase matrix corresponding to the current time step are used as inputs to the joint action determination model to obtain the joint action value of the current time step.

[0037] In one possible implementation, after adjusting the base station based on the second beamforming matrix and adjusting the reconfigurable smart surface based on the second phase matrix, the processing module is further configured to:

[0038] Based on the second beamforming matrix and the second phase matrix, the second state information of multiple communication links in the next time step is obtained;

[0039] By integrating the first state information, the second state information, the first beamforming matrix, the first phase matrix, the second beamforming matrix, and the second phase matrix, empirical data for the current time step is obtained.

[0040] Experience data is stored in an experience pool.

[0041] In one possible implementation, the processing module is further configured to:

[0042] Based on immediate rewards, the joint actions determine the objective function of the model at the current time step;

[0043] Based on the objective function, the first beamforming matrix, and the first phase matrix, the loss function of the joint action determination model at the current time step is determined.

[0044] Based on the objective function and loss function, the joint action determination model is trained using the gradient descent algorithm to obtain the joint action value at the current time step.

[0045] In one possible implementation, the processing module is specifically used for:

[0046] Based on the joint action value of the current time step, determine the beamforming network gradient and phase control network gradient corresponding to the current time step;

[0047] Based on the gradients of multiple communication links and beamforming network at the current time step, the second beamforming matrix for the next time step is obtained through the beamforming network.

[0048] Based on the gradients of multiple communication links and the phase control network at the current time step, the second phase matrix for the next time step is obtained through the phase control network.

[0049] In one possible implementation, the processing module is further configured to:

[0050] Based on the product of large-scale fading and small-scale fading of each communication link, an equivalent baseband channel is constructed for each communication link, resulting in multiple equivalent baseband channels. Among them, the equivalent baseband channels include direct equivalent baseband channels, indirect equivalent baseband channels, reconstructed equivalent baseband channels, direct eavesdropping equivalent baseband channels, and reconstructed eavesdropping equivalent baseband channels.

[0051] Based on the direct equivalent baseband channel, the indirect equivalent baseband channel, the reconstructed equivalent baseband channel, and the additive complex Gaussian noise corresponding to each user, the achievable transmission rate for each user is obtained.

[0052] Based on the indirect equivalent baseband channel, the direct eavesdropping equivalent baseband channel, the reconstructed eavesdropping equivalent baseband channel, and the additive complex Gaussian noise corresponding to the eavesdropper, the achievable eavesdropping rate corresponding to each eavesdropper is obtained.

[0053] Based on the achievable transmission rate and achievable eavesdropping rate, determine the achievable security rate for the current time step and obtain the instant reward corresponding to the current time step.

[0054] Thirdly, this application provides a transmission system for performing the methods described in the first aspect and / or various possible implementations of the first aspect.

[0055] Fourthly, this application provides an electronic device, including: a memory and a processor;

[0056] The memory stores instructions that the computer executes;

[0057] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0058] Fifthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible embodiments of the first aspect.

[0059] In a sixth aspect, this application provides a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0060] The intelligent anti-cooperative eavesdropping secure communication method, apparatus, device, and medium provided in this application acquire first state information, a first beamforming matrix, and a first phase matrix of multiple communication links at the current time step, enabling effective differentiation between legitimate users and eavesdroppers. Based on the first state information, the first beamforming matrix, and the first phase matrix, the joint action value of the current time step is obtained, which can guide the base station and reconfigurable smart surface to optimize their coordinated adjustment strategy, maximizing the overall performance of the transmission system. Based on the joint action value, the first beamforming matrix, and the first phase matrix, with maximizing the privacy transmission rate as the optimization objective, a second beamforming matrix and a second phase matrix for the next time step are obtained. This makes the adjustments of the base station and the reconfigurable smart surface more precise in the next time step, enhancing the directionality and concealment of communication, thereby reducing the likelihood of eavesdroppers successfully receiving signals and further improving the performance and stability of the transmission system. Adjusting the base station based on the second beamforming matrix and the reconfigurable smart surface based on the second phase matrix can enable the transmission system formed by the base station and the reconfigurable smart surface to have a better privacy transmission rate. This ensures that information is transmitted securely and effectively to legitimate users while preventing eavesdroppers from eavesdropping on the information, thus achieving a better anti-eavesdropping effect. This improves the performance of the communication link, realizes dynamic optimization and adaptive adjustment of the transmission system to prevent eavesdropping, and enhances the security and reliability of communication. Attached Figure Description

[0061] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0062] Figure 1 This is a schematic diagram of the transmission system provided in an embodiment of this application;

[0063] Figure 2 A flowchart illustrating the intelligent anti-collusive eavesdropping secure communication method provided in the embodiments of this application. Figure 1 ;

[0064] Figure 3 A flowchart illustrating the intelligent anti-collusive eavesdropping secure communication method provided in the embodiments of this application. Figure 2 ;

[0065] Figure 4 A schematic diagram of the structure of a transmission device based on a reconfigurable smart surface provided in an embodiment of this application;

[0066] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0067] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0068] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0069] Related technology 1, using AN injection to prevent eavesdropping, introduces interference signals into the base station's transmitted signals to disrupt the eavesdropping link. However, the interference effect of AN is limited when there are many eavesdroppers or the channel's degrees of freedom are restricted, and it may increase the base station's transmission power overhead. Related technology 2, using cooperative jamming (CJ) to prevent eavesdropping, involves legitimate nodes collaboratively transmitting interference signals to suppress eavesdropper reception. However, CJ relies on cooperation between nodes, and when eavesdroppers fully share information, the interference signal may be canceled out, leading to a decrease in security. Related technology 3, using beamforming and channel enhancement to prevent eavesdropping, reconfigurable smart surfaces dynamically adjust the phase matrix to reconfigure the wireless channel and enhance the signal strength of legitimate users. However, the suppression strategy for the eavesdropping link is insufficient, and there is a problem of not being able to effectively and promptly interfere with eavesdroppers. Related technology 4, using joint optimization of multiple reconfigurable smart surfaces to prevent eavesdropping, is difficult to cope with dynamic environments and collaborative attacks by multiple eavesdroppers. Therefore, there is an urgent need for a secure and effective transmission scheme based on reconfigurable smart surfaces to achieve a secure and effective anti-eavesdropping effect.

[0070] The intelligent anti-cooperative eavesdropping secure communication method provided in this application acquires first state information, a first beamforming matrix, and a first phase matrix of multiple communication links at the current time step, enabling effective differentiation between legitimate users and eavesdroppers. Based on the first state information, the first beamforming matrix, and the first phase matrix, the joint action value of the current time step is obtained, which can guide the base station and the reconfigurable smart surface to optimize their coordinated adjustment strategy, maximizing the overall performance of the transmission system. Based on the joint action value, the first beamforming matrix, and the first phase matrix, with maximizing the privacy transmission rate as the optimization objective, a second beamforming matrix and a second phase matrix are obtained for the next time step. This makes the adjustments of the base station and the reconfigurable smart surface more precise in the next time step, enhancing the directionality and concealment of communication, thereby reducing the likelihood of eavesdroppers successfully receiving signals and further improving the performance and stability of the transmission system. Adjusting the base station based on the second beamforming matrix and the reconfigurable smart surface based on the second phase matrix can enable the transmission system formed by the base station and the reconfigurable smart surface to have a better privacy transmission rate. This ensures that information is transmitted securely and effectively to legitimate users while preventing eavesdroppers from eavesdropping on the information, thus achieving a better anti-eavesdropping effect. This improves the performance of the communication link, realizes dynamic optimization and adaptive adjustment of the transmission system to prevent eavesdropping, and enhances the security and reliability of communication.

[0071] Figure 1 This is a schematic diagram of the transmission system provided in an embodiment of this application. Figure 1 The transmission system includes a base station 11 with N antennas, a reconfigurable smart surface 12 with M passive reflective elements, K single-antenna users 13, and L single-antenna eavesdroppers 14, wherein:

[0072] The communication links in the transmission system include the communication link from base station 11 directly to user 13, the communication link from base station 11 to eavesdropper 14, the communication link from base station 11 to reconfigurable smart surface 12, the communication link reflected from reconfigurable smart surface 12 to user 13, and the communication link reflected from reconfigurable smart surface 12 to eavesdropper 14.

[0073] Furthermore, during the operation of the transmission system in the above example, base station 11 sends confidential information to K users 13 within the same frequency band, using... This represents the user. At this point, let's assume... An eavesdropper 14, attempting to eavesdrop alone on information transmitted by base station 11, used... This represents eavesdropper 14. The transmitted signal at base station 11 can be represented as:

[0074]

[0075] in, This represents the independent Gaussian data symbol corresponding to the k-th user (13), further... follow ; This represents the beamforming matrix corresponding to base station 11. The beamforming matrix corresponding to base station 11 can be obtained through communication interaction with base station 11 in the transmission system.

[0076] For ease of calculation, the set of all precoding vectors in the beamforming matrix corresponding to base station 11 is represented as: ,and Meets power limits ,in, This is the maximum transmit power of base station 11.

[0077] By communicating with the reconfigurable smart surface 12 in the transmission system, the beamforming matrix corresponding to the reconfigurable smart surface 12 can be obtained. The phase matrix of the reconfigurable smart surface 12 can be expressed as:

[0078]

[0079] in , Indicates the first The reflection coefficient of each reflective element.

[0080] For example, suppose the reflection efficiency of the reflective element in the reconfigurable smart surface 12 is... Then the phase matrix of the reconfigurable smart surface 12 can be expressed as: .

[0081] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0082] Figure 2 A flowchart illustrating the intelligent anti-collusive eavesdropping secure communication method provided in the embodiments of this application. Figure 1 .like Figure 2 As shown, this method is applied to a transmission system composed of a base station and a reconfigurable smart surface, including:

[0083] S201. Obtain the first state information, first beamforming matrix, and first phase matrix of multiple communication links at the current time step. The communication links include the communication links of legitimate users and the communication links of eavesdroppers.

[0084] The current time step represents the specific time in the time series of the transmission system. Optionally, the current time step can be the current point in time or the current time period.

[0085] A communication link refers to a series of transmission paths and related equipment in a transmission system, connecting the transmitter and receiver, used to transmit signals. Through the communication link, legitimate users and eavesdroppers can be distinguished. This allows for the enhancement of signal strength for legitimate users when determining the second beamforming matrix and second phase matrix for the next time step, ensuring timely and effective signal reception. Simultaneously, it interferes with eavesdropping attempts, thereby improving communication security.

[0086] The first state information is used to describe the channel state and communication state of the communication link at the current time step.

[0087] In wireless communication, beamforming is a technique that adjusts the phase and amplitude of signals from individual antennas in an antenna array to create a high-gain beam in a specific direction. The first beamforming matrix describes the phase and amplitude adjustment parameters of the signals from each antenna in the base station's antenna array at the current time step.

[0088] Reconfigurable smart surfaces can optimize communication performance by adjusting the phase of their internal reflective elements, thereby altering the propagation characteristics of wireless signals. The first phase matrix describes the phase adjustment parameters of each reflective element in the reconfigurable smart surface within the current time step.

[0089] By using sensors and signal processing modules on the base station and the reconfigurable smart surface, information corresponding to multiple communication links at the current time step is monitored and collected in real time to obtain first state information. Optionally, the first state information includes at least one of the following: signal strength and channel state of each communication link. Simultaneously, the base station calculates the beamforming matrix for the current time step based on its antenna array configuration and signal processing algorithm, obtaining the first beamforming matrix for the current time step; the reconfigurable smart surface obtains the first phase matrix for the current time step based on its internal reflection unit control algorithm. By acquiring the first state information, the first beamforming matrix, and the first phase matrix for the current time step, the operating state of the transmission system in the current time step can be fully understood, enabling reasonable adjustments and decisions based on the operating state of the transmission system to improve communication quality and efficiency.

[0090] S202. Based on the first state information, the first beamforming matrix, and the first phase matrix, the joint action value of the current time step is obtained.

[0091] Joint action value is a comprehensive assessment of the benefits or performance improvements obtained from the current actions of the base station and the reconfigurable smart surface in a transmission system where the base station and the reconfigurable smart surface cooperate. Joint action value comprehensively considers the mutual influence and synergistic effects between the base station and the reconfigurable smart surface.

[0092] By employing deep learning, the first state information, the first beamforming matrix, and the first phase matrix are taken as input. A trained neural network model outputs estimated values ​​for different joint actions taken in the current state, selecting the highest value as the joint action value for the current time step. By determining the joint action value for the current time step, the base station and the reconfigurable smart surface can be guided to optimize their collaborative adjustment strategy, enabling their actions to maximize the overall performance of the transmission system.

[0093] S203. Based on the joint action value, the first beamforming matrix, and the first phase matrix, with the optimization objective of maximizing the privacy transmission rate, the second beamforming matrix and the second phase matrix for the next time step are obtained.

[0094] The second beamforming matrix refers to the ideal value of the beamforming matrix that the antenna array in the base station needs to achieve in the next time step in order to enhance the signal strength received by the user and interfere with eavesdroppers' eavesdropping on the communication link.

[0095] The second phase matrix refers to the ideal matrix that, in the next time step, requires the phase values ​​of each reflective unit within the reconstructable smart surface to be adjusted.

[0096] By using optimization algorithms with the goal of maximizing privacy transmission rate, and combining information based on joint action value, first beamforming matrix, and first phase matrix, the second beamforming matrix and second phase matrix for the next time step are obtained. This enables the base station and reconfigurable smart surface to adapt to changes in the transmission environment, make better adjustments, and further improve the performance and stability of the transmission system.

[0097] For example, the optimization problem, with the second beamforming matrix and the second phase matrix as optimization variables and maximizing the privacy transmission rate as the optimization objective, can be expressed as:

[0098]

[0099] in, This represents the precoding vector corresponding to the beamforming matrix; Indicates the phase parameter; No. The transmission rate at which each user receives confidential messages; Indicates L eavesdroppers. The first eavesdropper received the first The achievable eavesdropping rate of a user's confidential messages. Considering the worst-case scenario where information can be completely shared and coordinated among eavesdroppers, the optimization objective is set as: the user's confidentiality rate minus the information rate of all eavesdroppers, rather than the maximum rate at which eavesdroppers can eavesdrop. This improves the effectiveness of anti-eavesdropping measures while enhancing user signal transmission and ensuring the rate of transmission to the user.

[0100] Furthermore, maximizing the long-term cumulative expected return can also be expressed as:

[0101]

[0102] in, This represents the k-th vector of the precoding vector corresponding to the mid-beamforming matrix; This is the maximum transmit power of the base station; Represents the N reflective elements of a reconfigurable smart surface. The reflection coefficient of each reflective element.

[0103] S204, Adjust the base station based on the second beamforming matrix, and adjust the reconfigurable smart surface based on the second phase matrix.

[0104] The base station, through its internal signal processing and control module, adjusts the signal phase and amplitude of each antenna in the antenna array within the base station according to the calculated second beamforming matrix to achieve a specific beamforming effect. The reconfigurable smart surface, through its internal control circuitry, adjusts the phase of each reflective element within the reconfigurable smart surface according to the second phase matrix, thereby altering the propagation path and characteristics of the wireless signal.

[0105] By adjusting the base station according to the second beamforming matrix and the reconfigurable smart surface according to the second phase matrix, the transmission system formed by the base station and the reconfigurable smart surface can achieve a better privacy transmission rate. This ensures that information is transmitted securely and effectively to legitimate users while preventing eavesdroppers from intercepting the information, thus achieving a better anti-eavesdropping effect. This improves the performance of the communication link, enables dynamic optimization and adaptive adjustment of the transmission system to prevent eavesdropping, and enhances the security and reliability of communication.

[0106] Furthermore, by adjusting the base station and the reconfigurable smart surface at multiple time steps, the transmission system can achieve increasingly higher transmission rates and better anti-eavesdropping effects.

[0107] The intelligent anti-cooperative eavesdropping security communication method provided in this application obtains the first state information, first beamforming matrix, and first phase matrix of multiple communication links at the current time step to effectively distinguish between legitimate users and eavesdroppers. Based on the first state information, first beamforming matrix, and first phase matrix, the joint action value of the current time step is obtained, which can guide the base station and reconfigurable smart surface to optimize their coordinated adjustment strategy, enabling the actions of the base station and reconfigurable smart surface to maximize the overall performance of the transmission system. Based on the joint action value, first beamforming matrix, and first phase matrix, with maximizing the privacy transmission rate as the optimization objective, the second beamforming matrix and second phase matrix for the next time step are obtained. This makes the adjustments of the base station and reconfigurable smart surface more precise in the next time step, enhancing the directionality and concealment of communication, thereby reducing the possibility of eavesdroppers successfully receiving signals and further improving the performance and stability of the transmission system. Adjusting the base station based on the second beamforming matrix and the reconfigurable smart surface based on the second phase matrix can enable the transmission system formed by the base station and the reconfigurable smart surface to have a better privacy transmission rate. This ensures that information is transmitted securely and effectively to legitimate users while preventing eavesdroppers from eavesdropping on the information, thus achieving a better anti-eavesdropping effect. This improves the performance of the communication link, realizes dynamic optimization and adaptive adjustment of the transmission system to prevent eavesdropping, and enhances the security and reliability of communication.

[0108] Figure 3 A flowchart illustrating the intelligent anti-collusive eavesdropping secure communication method provided in the embodiments of this application. Figure 2 .like Figure 3 As shown, in this embodiment... Figure 2 Based on the embodiments, the intelligent anti-collusive eavesdropping security communication method is described in detail. The method includes:

[0109] In one possible implementation, step S202 may further include:

[0110] S2021. Based on the first state information, determine the achievable security rate of the current time step and obtain the instant reward corresponding to the current time step.

[0111] In the field of wireless communication security, the achievable confidentiality rate refers to the maximum reliable data transmission rate that legitimate communicating parties can achieve while ensuring the security of information transmission in a legitimate communication link.

[0112] Immediate reward is the feedback signal that the base station and the reconfigurable smart surface collaborative system immediately receive from the environment after taking an action at the current time step. Immediate reward measures the direct benefit brought by the action in the current state. For example, the achievable confidentiality rate can be used as the immediate reward, reflecting the secure communication performance of the transmission system under the action corresponding to the current time step.

[0113] First, the communication links of legitimate users and eavesdroppers are distinguished, obtaining the first state information corresponding to the legitimate user's communication link and the eavesdropper's communication link. Then, the transmission rate of the legitimate communication link at the current time step is calculated based on the first state information of the legitimate user's communication link; the information rate that the eavesdropping link might intercept at the current time step is calculated based on the first state information of the eavesdropper's communication link. Finally, the achievable confidentiality rate is obtained by calculating the difference between the transmission rate of the legitimate communication link and the information rate that the eavesdropping link might intercept, and this achievable confidentiality rate is used as the immediate reward for the current time step.

[0114] For example, within the current time step, the confidentiality rate between the k-th legitimate user and the L eavesdroppers... Defined as:

[0115]

[0116] in, No. The eavesdropping rate at which a legitimate user can receive confidential messages; Indicates L eavesdroppers. The first eavesdropper received the first The achievable rate of eavesdropping on confidential messages of a legitimate user.

[0117] Then, through the confidentiality rate Power consumption when working with base stations and reconfigurable smart surfaces is rewarded instantly. .

[0118] Immediate rewards based on achievable confidentiality rates enable transmission systems to pay more attention to communication security during reinforcement learning, guiding them to adopt action strategies that improve confidentiality rates, thereby optimizing the security performance of the transmission system.

[0119] S2022. Using the immediate reward, the first beamforming matrix, and the first phase matrix corresponding to the current time step as inputs to the joint action determination model, the joint action value of the current time step is obtained.

[0120] A joint action determination model is a mathematical model used to evaluate the joint action value corresponding to the current joint action of the base station and the reconfigurable smart surface based on the immediate reward, the first beamforming matrix, and the first phase matrix corresponding to the current time step. Optionally, the joint action determination model can be a reinforcement learning model. The joint action value is the total future gain of the transmission system under the current communication state and possible cooperative adjustment actions.

[0121] The immediate reward, first beamforming matrix, and first phase matrix corresponding to the current time step are substituted into the joint action determination model. By iteratively solving the joint action model, the joint action value for the current time step is obtained. Accurately determining the joint action value through the joint action determination model can guide the transmission system in selecting appropriate cooperative adjustment strategies, making the actions of the base station and reconfigurable smart surface more rational, and further improving the overall performance of the transmission system.

[0122] For example, the value of joint actions It can be represented as:

[0123]

[0124] in, Indicates the first state information; Represents the joint action of the base station and the reconfigurable smart surface; This represents the first beamforming matrix; Represents the first phase matrix; This represents the real-time parameters of the joint action determination model at the current time step, used to represent the long-term expected security rate under the beamforming matrix and phase matrix.

[0125] In one possible implementation, after step S204, the method further includes:

[0126] S205. Obtain the second state information of multiple communication links in the next time step.

[0127] After adjusting the base station and the reconfigurable smart surface, the second state information of multiple communication links after the adjustment is obtained, and the second state information of multiple communication links in the next time step is obtained.

[0128] S206. By integrating the first state information, the second state information, the first beamforming matrix, the first phase matrix, the second beamforming matrix, and the second phase matrix, empirical data for the current time step is obtained.

[0129] By integrating the first state information, the second state information, the first beamforming matrix, the first phase matrix, the second beamforming matrix, and the second phase matrix, empirical data for the current time step is obtained.

[0130] By integrating the first state information Second state information First beamforming matrix First phase matrix Second beamforming matrix Second phase matrix The empirical data for the current time step is as follows:

[0131]

[0132] Optionally, the experience data may also include the immediate reward corresponding to the current time step.

[0133] For example, the first state information of multiple communication links at the current time step. for:

[0134]

[0135] in, This indicates the direct link from the base station to the user; This indicates an indirect link from the base station to the reconfigurable smart surface; This represents the reconfiguration link from the reconfigurable smart surface to the user; Indicates the direct eavesdropping link from the base station to the eavesdropper and This indicates a reconfigurable eavesdropping link from a reconfigurable smart surface to an eavesdropper.

[0136] For example, the first beamforming matrix and the first phase matrix By integrating, a joint action representation of the current time step base station and the reconfigurable smart surface is obtained. By integrating joint actions to represent First state information Second state information and instant rewards The empirical data for the current time step is as follows: .

[0137] S207. Store the experience data in the experience pool.

[0138] An experience pool is a data structure used to store experiential data collected by an agent during its interaction with its environment. An experience pool is typically a fixed-size queue or list that stores experiential data according to certain rules. The main function of the experience pool is to break the temporal correlation between experiential data, making the training samples more independent and identically distributed, thereby improving the stability and convergence of reinforcement learning algorithms.

[0139] For example, an experience pool data structure of appropriate size is created, and the experience data of the current time step is added to the experience pool according to the storage rules of the experience pool. If the experience pool is full, an old experience data point in the experience pool is replaced according to the set replacement strategy to ensure that the experience pool always maintains a certain amount of experience data available for training. Optionally, the replacement strategy includes any one of the following strategies: random replacement or first-in-first-out replacement.

[0140] The empirical data in the experience pool provides rich training samples for the joint action determination model. This empirical data includes decision-making and state evolution information of the base station and reconfigurable smart surface under different states within the transmission system. This helps the algorithm learn better decision-making strategies, enabling the transmission system to make more reasonable adjustments based on the current state, thereby improving communication performance and security.

[0141] For example, when the amount of experience data in the experience pool reaches a certain amount, the parameters in the joint action determination model are updated based on the data in the experience pool, so that the joint action determination model can dynamically calculate the value of joint actions according to the transmission requirements of the transmission system within a specific time period, ensuring that the value of joint actions can better fit the actual transmission environment.

[0142] In one possible implementation, the joint action determination model for the current time step is obtained in the following way:

[0143] Step A: Based on immediate rewards, determine the objective function of the joint action model at the current time step.

[0144] First, the joint action determination model is retrieved from the storage unit. Optionally, the joint action determination model from the previous time step can be obtained by retrieving its parameters. Acquiring the joint action determination model leverages historical decision-making experience, avoiding the need to build a model from scratch and saving computational resources and time. Simultaneously, the objective function of the joint action model is adjusted through immediate rewards, ensuring that the joint action model has a certain correlation with the system state at the current time step. This provides valuable reference for decision-making at the current time step, improving the accuracy and efficiency of decision-making.

[0145] Instant rewards based on the current time step Desired beamforming matrix Desired phase matrix and expected joint action parameters The objective function of the joint action determination model at the current time step is obtained. for:

[0146]

[0147] Among them, the desired beamforming matrix is ​​the beamforming parameter matrix expected to be used to achieve optimal communication performance under ideal conditions. The desired phase matrix is ​​the antenna phase adjustment parameter matrix expected to be used to achieve optimal communication performance under ideal conditions. The desired joint action parameters are the set of model parameters for determining the joint actions expected to be used to achieve optimal communication performance under ideal conditions.

[0148] For example, considering the reflection and beaming effects corresponding to the actions of the base station and the reconfigurable smart surface in future time steps, a joint action determination model can be constructed based on a multi-step objective function to determine the objective function of the model in the current time step. It can be represented as:

[0149]

[0150] in, Indicates an immediate reward for the next step; This represents the immediate reward for the next n-1 steps. Representation and weighting coefficients.

[0151] Step B: Based on the objective function, the first beamforming matrix, and the first phase matrix, the loss function of the joint action model at the current time step is determined by minimizing the temporal difference.

[0152] According to the objective function First beamforming matrix and the first phase matrix By minimizing the temporal difference, the joint action is determined to define the loss function of the model at the current time step. for:

[0153]

[0154] Where E represents the objective function and joint action value function Expectations between them; This indicates the instantaneous channel information and system configuration parameters at the current time step; This indicates that the joint actions at the current time step determine the real-time parameters of the model.

[0155] Step C: Based on the objective function and loss function, train the joint action determination model using the gradient descent algorithm to obtain the joint action value at the current time step.

[0156] The desired joint action parameters are obtained using the gradient descent algorithm. Optimization is performed to obtain the real-time parameters of the joint action determination model corresponding to the current time step. The process can be represented as:

[0157]

[0158] in, Indicates the learning rate; Represents the loss function Regarding real-time parameters The gradient.

[0159] By training the joint action determination model based on the objective function and loss function using the gradient descent algorithm, the joint action value at the current time step is obtained. This enables the joint action determination model to continuously adapt to changes in the transmission system state and uncertainties in the communication environment, allowing the transmission system to more accurately select the appropriate second beamforming matrix and second phase matrix based on the joint action value.

[0160] In one possible implementation, step S203 may further include:

[0161] S2031. Based on the joint action value of the current time step, determine the beamforming network gradient and phase control network gradient corresponding to the current time step.

[0162] Beamforming network gradient refers to the gradient of beamforming network parameters with respect to the joint action value. It represents the rate of change of the joint action value with respect to the beamforming network parameters. By using the beamforming network gradient, the beamforming network parameters can be effectively adjusted to adjust the joint action value of beamforming in joint actions.

[0163] The phase control network gradient is used to generate the gradient of the phase control network parameters with respect to the value of the joint action, and is used to indicate how to adjust the parameters within the phase control network.

[0164] Based on the joint action value at the current time step, the beamforming network gradient of the beamforming network corresponding to the current time step and the phase control network gradient of the phase control network corresponding to the current time step are calculated respectively. This clarifies the adjustment direction of the parameters within the beamforming network and the phase control network, enabling them to optimize towards improving the joint action value. Simultaneously, it ensures that the beamforming network and the phase control network can be jointly updated under consistent joint action values, thereby achieving global coordination between base station beamforming and reconfigurable intelligent surface phase control.

[0165] For example, based on the joint action value at the current time step, the beamforming network gradient at the current time step is obtained through the beamforming network gradient update formula. :

[0166]

[0167] in, Describe the policy function under the conditions of the current time step. Regarding beamforming network parameters The gradient; Represents the value function of joint actions Beamforming matrix for the current time step The gradient.

[0168] For example, based on the joint action value at the current time step, the gradient of the phase control network at the current time step is obtained through the gradient update formula of the phase control network. :

[0169]

[0170] in, Describe the policy function under the conditions of the current time step. Regarding phase control network parameters The gradient; Represents the value function of joint actions Phase matrix at the current time step gradient.

[0171] S2032. Based on the gradients of multiple communication links and beamforming network at the current time step, the beamforming matrix at the current time step is updated through the beamforming network to obtain the second beamforming matrix for the next time step.

[0172] Multiple communication links at the current time step, along with the beamforming network gradients, are input into the beamforming network. The adjustment direction and magnitude of the beamforming network parameters are determined based on the beamforming network gradients, resulting in an updated beamforming network. Subsequently, the updated beamforming network is used to obtain the second beamforming matrix for the next time step, enabling the second beamforming matrix to more accurately adapt to changes in communication links and improve signal transmission quality and coverage.

[0173] For example, the second beamforming matrix for:

[0174]

[0175] in, This represents the beamforming network parameters determined based on the beamforming network gradient. Represents multiple communication links at the current time step; This represents Gaussian exploration noise.

[0176] S2033. Based on the gradients of multiple communication links and the phase control network at the current time step, the phase matrix at the current time step is updated through the phase control network to obtain the second phase matrix for the next time step.

[0177] The gradients of multiple communication links at the current time step, along with the gradients of the phase control network, are input into the phase control network. The adjustment direction and magnitude of the phase control network parameters are determined based on the gradients, resulting in an updated phase control network. Then, the updated phase control network is used to obtain the second phase matrix for the next time step, enabling the second phase matrix to more accurately adapt to changes in communication links and improve signal transmission quality and coverage.

[0178] For example, the second phase matrix for:

[0179]

[0180] in, The gradient of the phase control network determines the parameters of the phase control network. Represents multiple communication links at the current time step; This represents Gaussian exploration noise.

[0181] In one possible implementation, step S2021 may further include:

[0182] Step A: Based on the product of large-scale fading and small-scale fading of each communication link, construct the equivalent baseband channel corresponding to each communication link to obtain multiple equivalent baseband channels; among them, the equivalent baseband channels include direct equivalent baseband channels, indirect equivalent baseband channels, reconstructed equivalent baseband channels, direct eavesdropping equivalent baseband channels, and reconstructed eavesdropping equivalent baseband channels.

[0183] Based on the product of large-scale and small-scale fading for each communication link, equivalent baseband channels are constructed for each communication link, thus obtaining the equivalent baseband channel for each communication link. Communication links include: base station to user communication links, base station to reconfigurable smart surface communication links, reconfigurable smart surface to user communication links, base station to eavesdropper communication links, and reconfigurable smart surface to eavesdropper communication links. Equivalent baseband channels include: direct equivalent baseband channels for base station to user communication links, indirect equivalent baseband channels for base station to reconfigurable smart surface communication links, reconfigurable equivalent baseband channels for reconfigurable smart surface to user communication links, direct eavesdropping equivalent baseband channels for base station to eavesdropper communication links, and reconfigurable eavesdropping equivalent baseband channels for reconfigurable smart surface to eavesdropper communication links.

[0184] Large-scale fading is caused by changes in signal strength due to macroscopic factors such as terrain and buildings in the communication link. Large-scale fading is a slow process and is mainly related to factors such as the distance the signal travels and the distribution of obstacles. Small-scale fading, on the other hand, is caused by changes in the instantaneous amplitude and phase of the signal due to multipath propagation effects. Within a short period, the signal undergoes rapid changes.

[0185] Optionally, large-scale fading can be described using a path loss model.

[0186] For example, taking the communication link from base station to user as an example, the large-scale fading corresponding to the communication link from base station to user is described by the path loss model, and the following is obtained:

[0187]

[0188] in, This represents the path loss from the base station to a distance d. Distance from base station The reference path loss is n, where n is the path loss exponent.

[0189] Optionally, the non-line-of-sight component of the communication link is obtained based on the effective scatterer and the complex gain of the effective scatterer within the communication link; the line-of-sight component of the communication link is obtained based on the transmit antenna gain, receive antenna gain, wavelength, length of the communication link, and free-space propagation delay within the communication link; and the ratio of the non-line-of-sight component to the line-of-sight component of the communication link is determined by the Rice factor to obtain the small-scale fading of the communication link.

[0190] For example, taking a base station to user communication link as an example, the non-line-of-sight component of the communication link is obtained based on the effective scatterers and the complex gain of the effective scatterers within the communication link. for:

[0191]

[0192] in, Indicates the number of effective scatterers; Represents complex gain. phase Evenly distributed in In the above, j represents the j-th user.

[0193] Based on the transmit antenna gain, receive antenna gain, wavelength, link length, and free-space propagation delay within the communication link, the line-of-sight component of the communication link is obtained. for:

[0194]

[0195] in, Indicates the transmit antenna gain; Indicates the receiver antenna gain; Table Indicate wavelength; Indicates the length of the communication link and This represents the free space propagation delay for the j-th user.

[0196] The ratio of the non-line-of-sight component to the line-of-sight component in a communication link is determined by the Rice factor, thus obtaining the small-scale fading of the communication link. for:

[0197]

[0198] in, Represents the Rice factor; represents the non-line-of-sight component. And represents the line-of-sight component. .

[0199] Therefore, large-scale fading based on the base station to user communication link and small-scale fading The product of these terms yields the direct equivalent baseband channel corresponding to the communication link from the base station to the user: .

[0200] Step B: Based on the direct equivalent baseband channel, the indirect equivalent baseband channel, the reconstructed equivalent baseband channel, and the additive complex Gaussian noise corresponding to the user, obtain the achievable transmission rate for each user.

[0201] Based on the direct equivalent baseband channel, the indirect equivalent baseband channel, the reconstructed equivalent baseband channel, and the additive complex Gaussian noise corresponding to the user, the security signal received by each user is obtained;

[0202] Based on the security signal received by each user, the power of interference from other users to each user is determined, resulting in the corresponding interference power for each user. The absolute value of the security signal received by each user is squared to obtain the signal power received by each user. The signal-to-noise ratio (SNR) and signal-to-interference-plus-noise ratio (SINR) for each user are obtained based on the ratio between the received signal power and the corresponding interference power. Based on the SNR and SINR, the achievable transmission rate for each user is determined using the transmission rate determination formula. By calculating the achievable transmission rate for each user, the transmission system's ability to provide data transmission services can be evaluated, which helps optimize the resource allocation of the transmission system.

[0203] For example, taking the communication link from base station to user as an example, based on the direct equivalent baseband channel, the indirect equivalent baseband channel, the reconstructed equivalent baseband channel, and the additive complex Gaussian noise corresponding to the user, the security signal received by the k-th user is obtained. for:

[0204]

[0205] in, This represents the indirect equivalent baseband channel from the base station to the reconfigurable smart surface; M represents the number of passive reflective elements in the reconfigurable smart surface; and N represents the number of antennas at the base station. This represents the direct equivalent baseband channel from the base station to the k-th user; This represents the reconfigurable equivalent baseband channel from the reconfigurable smart surface to the k-th user; This represents the additive complex Gaussian noise at the k-th user receiver. It follows a mean of 0 and a variance of . The complex Gaussian distribution.

[0206] Based on the security signals received by each user, the power of interference from other users to each user is determined, thus obtaining the interference power corresponding to the k-th user. for:

[0207]

[0208] The security signal received by the k-th user The absolute value of the signal is squared to obtain the signal power received by the k-th user. for:

[0209]

[0210] Based on the signal power received by the k-th user Interference power corresponding to the kth user The ratio between them yields the signal-to-noise ratio (SNR) and signal-to-interference-plus-noise ratio (SINNR) for the k-th user. for:

[0211]

[0212] Based on each user's signal-to-noise ratio and signal-to-interference-plus-noise ratio The achievable transmission rate for the k-th user can be obtained using the transmission rate determination formula. for:

[0213]

[0214] Step C: Based on the indirect equivalent baseband channel, the direct eavesdropping equivalent baseband channel, the reconstructed eavesdropping equivalent baseband channel and the additive complex Gaussian noise corresponding to the eavesdropper, obtain the achievable eavesdropping rate corresponding to each eavesdropper.

[0215] Based on the indirect equivalent baseband channel, the direct eavesdropping equivalent baseband channel, the reconstructed eavesdropping equivalent baseband channel, and the additive complex Gaussian noise corresponding to each eavesdropper, the eavesdropping signal received by each eavesdropper is determined. Based on the eavesdropping signal received by each eavesdropper, the power of interference from other eavesdroppers is determined, obtaining the corresponding interference power for each eavesdropper. The absolute value of the eavesdropping signal received by each eavesdropper is squared to obtain the signal power received by each eavesdropper. Based on the ratio between the signal power received by each eavesdropper and the corresponding interference power, the signal-to-noise ratio (SNR) and signal-to-interference-plus-noise ratio (SINR) of each eavesdropper are obtained. Based on the SNR and SINR of each eavesdropper, the achievable eavesdropping rate for each eavesdropper is obtained using the transmission rate determination formula. Calculating the achievable eavesdropping rate helps assess the security of the transmission system. By understanding the rate at which eavesdroppers can steal information, corresponding security measures can be taken to reduce the risk of information leakage.

[0216] For example, taking the communication link from base station to user as an example, the eavesdropping signal received by the l-th eavesdropper is determined based on the indirect equivalent baseband channel, the direct eavesdropping equivalent baseband channel, the reconstructed eavesdropping equivalent baseband channel and the additive complex Gaussian noise corresponding to the eavesdropper.

[0217]

[0218] in, This represents the indirect equivalent baseband channel from the base station to the reconfigurable smart surface; This represents the equivalent baseband channel for direct eavesdropping from the base station to the l-th eavesdropper among L eavesdroppers; This represents the equivalent baseband channel for reconfigurable eavesdropping from the reconfigurable smart surface to the l-th user; Let the additive complex Gaussian noise of the l-th eavesdropper be represented. It follows a mean of 0 and a variance of . The complex Gaussian distribution.

[0219] Based on the eavesdropping signals received by each eavesdropper Determine the power of interference experienced by each eavesdropper from other eavesdroppers, and obtain the interference power corresponding to the l-th eavesdropper. for:

[0220]

[0221] The eavesdropping signal received by the l-th eavesdropper The absolute value of the signal is squared to obtain the signal power received by each of the l-th eavesdroppers. for:

[0222]

[0223] Based on the signal power received by the k-th eavesdropper Interference power corresponding to each eavesdropper The ratio between them yields the signal-to-noise ratio (SNR) and signal-to-interference-plus-noise ratio (SINNR) of the l-th eavesdropper. for:

[0224]

[0225] Based on the signal-to-noise ratio and signal-to-interference-noise ratio of the l-th eavesdropper By using the transmission rate determination formula, the achievable eavesdropping rate for the l-th eavesdropper against the k-th user can be obtained. for:

[0226]

[0227] Step D: Based on the achievable transmission rate and achievable eavesdropping rate, determine the achievable security rate for the current time step and obtain the instant reward corresponding to the current time step.

[0228] The achievable security rate for the current time step is determined by subtracting the achievable eavesdropping rate of all eavesdroppers from the achievable transmission rate of each user, thus obtaining the instant reward corresponding to the current time step.

[0229] For example, the achievable security rate can be expressed as:

[0230]

[0231] in, No. Each user receives the achievable transmission rate; Indicates L eavesdroppers. The first eavesdropper received the first The achievable rate of eavesdropping on a user's confidential messages.

[0232] Figure 4 This is a schematic diagram of the structure of a transmission device based on a reconfigurable smart surface, provided in an embodiment of this application. Figure 4 As shown, the transmission device 40 based on a reconfigurable smart surface provided in this embodiment is deployed in a transmission system composed of a base station and a reconfigurable smart surface, including:

[0233] The acquisition module 401 is used to acquire the first state information, the first beamforming matrix and the first phase matrix of multiple communication links at the current time step. The communication links include the communication links of legitimate users and the communication links of eavesdroppers.

[0234] The processing module 402 is used to obtain the joint action value of the current time step based on the first state information, the first beamforming matrix, and the first phase matrix; based on the joint action value, the first beamforming matrix, and the first phase matrix, and with the optimization objective of maximizing the privacy transmission rate, obtain the second beamforming matrix and the second phase matrix of the next time step; adjust the base station based on the second beamforming matrix, and adjust the reconfigurable smart surface based on the second phase matrix.

[0235] In one possible implementation, the processing module 402 is specifically used for:

[0236] Based on the first state information, determine the achievable confidentiality rate at the current time step and obtain the immediate reward corresponding to the current time step;

[0237] The immediate reward, the first beamforming matrix, and the first phase matrix corresponding to the current time step are used as inputs to the joint action determination model to obtain the joint action value of the current time step.

[0238] In one possible implementation, after adjusting the base station based on the second beamforming matrix and adjusting the reconfigurable smart surface based on the second phase matrix, the processing module 402 is further configured to:

[0239] Based on the second beamforming matrix and the second phase matrix, the second state information of multiple communication links in the next time step is obtained;

[0240] By integrating the first state information, the second state information, the first beamforming matrix, the first phase matrix, the second beamforming matrix, and the second phase matrix, empirical data for the current time step is obtained.

[0241] Experience data is stored in an experience pool.

[0242] In one possible implementation, the processing module 402 is further configured to:

[0243] Based on immediate rewards, the joint actions determine the objective function of the model at the current time step;

[0244] Based on the objective function, the first beamforming matrix, and the first phase matrix, the loss function of the joint action determination model at the current time step is determined.

[0245] Based on the objective function and loss function, the joint action determination model is trained using the gradient descent algorithm to obtain the joint action value at the current time step.

[0246] In one possible implementation, the processing module 402 is specifically used for:

[0247] Based on the joint action value of the current time step, determine the beamforming network gradient and phase control network gradient corresponding to the current time step;

[0248] Based on the gradients of multiple communication links and beamforming network at the current time step, the second beamforming matrix for the next time step is obtained through the beamforming network.

[0249] Based on the gradients of multiple communication links and the phase control network at the current time step, the second phase matrix for the next time step is obtained through the phase control network.

[0250] In one possible implementation, the processing module 402 is further configured to:

[0251] Based on the product of large-scale fading and small-scale fading of each communication link, an equivalent baseband channel is constructed for each communication link, resulting in multiple equivalent baseband channels. Among them, the equivalent baseband channels include direct equivalent baseband channels, indirect equivalent baseband channels, reconstructed equivalent baseband channels, direct eavesdropping equivalent baseband channels, and reconstructed eavesdropping equivalent baseband channels.

[0252] Based on the direct equivalent baseband channel, the indirect equivalent baseband channel, the reconstructed equivalent baseband channel, and the additive complex Gaussian noise corresponding to each user, the achievable transmission rate for each user is obtained.

[0253] Based on the indirect equivalent baseband channel, the direct eavesdropping equivalent baseband channel, the reconstructed eavesdropping equivalent baseband channel, and the additive complex Gaussian noise corresponding to the eavesdropper, the achievable eavesdropping rate corresponding to each eavesdropper is obtained.

[0254] Based on the achievable transmission rate and achievable eavesdropping rate, determine the achievable security rate for the current time step and obtain the instant reward corresponding to the current time step.

[0255] The transmission device based on a reconfigurable smart surface provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0256] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 As shown, the electronic device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus 504.

[0257] In a specific implementation, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to perform the above-described method.

[0258] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0259] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0260] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0261] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings of this application's embodiments are not limited to only one bus or one type of bus.

[0262] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0263] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed, implement any of the methods described above.

[0264] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random-Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0265] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an application-specific integrated circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0266] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0267] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0268] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0269] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0270] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0271] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A smart anti-collusive eavesdropping secure communication method, characterized in that, Transmission systems composed of base stations and reconfigurable smart surfaces include: Acquire the first state information, first beamforming matrix, and first phase matrix of multiple communication links at the current time step, wherein the communication links include the communication links of legitimate users and the communication links of eavesdroppers; Based on the first state information, the first beamforming matrix, and the first phase matrix, the joint action value of the current time step is obtained; Based on the combined action value, the first beamforming matrix, and the first phase matrix, with the optimization objective of maximizing the privacy transmission rate, the second beamforming matrix and the second phase matrix for the next time step are obtained. The base station is adjusted based on the second beamforming matrix, and the reconfigurable smart surface is adjusted based on the second phase matrix.

2. The method according to claim 1, characterized in that, The step of obtaining the joint action value of the current time step based on the first state information, the first beamforming matrix, and the first phase matrix, with the optimization objective of maximizing the privacy transmission rate, includes: Based on the first state information, the achievable security rate of the current time step is determined, and the instant reward corresponding to the current time step is obtained; The immediate reward corresponding to the current time step, the first beamforming matrix, and the first phase matrix are used as inputs to the joint action determination model to obtain the joint action value of the current time step.

3. The method according to claim 2, characterized in that, After adjusting the base station based on the second beamforming matrix and adjusting the reconfigurable smart surface based on the second phase matrix, the method further includes: Based on the second beamforming matrix and the second phase matrix, the second state information of multiple communication links in the next time step is obtained; By integrating the first state information, the second state information, the first beamforming matrix, the first phase matrix, the second beamforming matrix, and the second phase matrix, empirical data for the current time step is obtained; The experience data is stored in the experience pool.

4. The method according to claim 3, characterized in that, The method uses the immediate reward corresponding to the current time step, the first beamforming matrix, and the first phase matrix as inputs to the joint action determination model to obtain the joint action value at the current time step: Based on immediate rewards, determine the objective function of the joint action model at the current time step; Based on the objective function, the first beamforming matrix, and the first phase matrix, the loss function of the joint action determination model at the current time step is determined; Based on the objective function and the loss function, the joint action determination model is trained using the gradient descent algorithm to obtain the joint action value at the current time step.

5. The method according to claim 1, characterized in that, The process of obtaining the second beamforming matrix and the second phase matrix for the next time step based on the combined action value, the first beamforming matrix, and the first phase matrix includes: Based on the joint action value of the current time step, determine the beamforming network gradient and phase control network gradient corresponding to the current time step; Based on the multiple communication links and the beamforming network gradient at the current time step, the second beamforming matrix for the next time step is obtained through the beamforming network. Based on the gradients of the multiple communication links and the phase control network at the current time step, the second phase matrix for the next time step is obtained through the phase control network.

6. The method according to claim 2, characterized in that, The process of determining the achievable security rate for the current time step based on the multiple communication links at the current time step, and obtaining the instant reward corresponding to the current time step, includes: Based on the product of large-scale fading and small-scale fading of each communication link, an equivalent baseband channel is constructed for each communication link, resulting in multiple equivalent baseband channels; wherein, the equivalent baseband channels include direct equivalent baseband channels, indirect equivalent baseband channels, reconstructed equivalent baseband channels, direct eavesdropping equivalent baseband channels, and reconstructed eavesdropping equivalent baseband channels. Based on the direct equivalent baseband channel, the indirect equivalent baseband channel, the reconstructed equivalent baseband channel, and the additive complex Gaussian noise corresponding to the user, the achievable transmission rate corresponding to each user is obtained; Based on the indirect equivalent baseband channel, the direct eavesdropping equivalent baseband channel, the reconstructed eavesdropping equivalent baseband channel, and the additive complex Gaussian noise corresponding to the eavesdropper, the achievable eavesdropping rate corresponding to each eavesdropper is obtained. Based on the achievable transmission rate and the achievable eavesdropping rate, the achievable security rate for the current time step is determined, and the instant reward corresponding to the current time step is obtained.

7. A transmission device based on a reconfigurable smart surface, characterized in that, Deployed in a transmission system consisting of a base station and a reconfigurable smart surface, including: The acquisition module is used to acquire the first state information, the first beamforming matrix, and the first phase matrix of multiple communication links at the current time step, wherein the communication links include the communication links of legitimate users and the communication links of eavesdroppers; The processing module is configured to: obtain the joint action value of the current time step based on the first state information, the first beamforming matrix, and the first phase matrix; obtain the second beamforming matrix and the second phase matrix of the next time step based on the joint action value, the first beamforming matrix, and the first phase matrix, with the optimization objective of maximizing the privacy transmission rate; adjust the base station based on the second beamforming matrix; and adjust the reconfigurable smart surface based on the second phase matrix.

8. A transmission system, characterized in that, Used to perform the method according to any one of claims 1-6.

9. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed, are used to implement the method as described in any one of claims 1-6.

Images