An omni-directional intelligent surface assisted physical layer security transmission method

Abstract

The application provides an omnidirectional intelligent surface assisted physical layer security transmission method, by introducing an intelligent omnidirectional surface into a downlink MISO wireless communication network, each non-active user in each system has a potential eavesdropping risk, by combining a transmitting beamforming vector of a transmitting base station and a passive beamforming vector at the intelligent omnidirectional surface, a wireless transmission channel can be reconstructed, a legal receiving channel can be enhanced, an eavesdropping channel can be weakened, and the system secrecy capacity can be improved; in addition, by calculating the lower bound of the system secrecy and rate, as long as the transmitting information is transmitted at a rate not higher than the secrecy and rate, the stability and safety of the transmission can be ensured from the information theory level.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication, specifically relating to an omnidirectional intelligent surface-assisted physical layer secure transmission method. Background Technology

[0002] In recent years, with the rapid development of technologies such as the Internet of Things, artificial intelligence, and mobile communications, communication technology has undergone continuous evolution through multiple generations from 2G and 3G to 5G. Today, with the official commercialization of fifth-generation (5G) mobile communication technology, a massive number of mobile devices need to access networks to seek high-quality communication services, and the types of communication services required are also rapidly increasing, such as... Figure 1 The demonstration showcased some application scenarios. The three main application scenarios for 5G technology are Ultra-Reliable and Low-Latency Communication (URLLC), Massive Machine-Type Communication (mMTC), and Enhanced Mobile Broadband (eMBB).

[0003] While striving to obtain greater bandwidth resources, higher channel capacity, higher communication rates, and lower communication latency, the privacy and security issues inherent in 5G communication networks also deserve attention. Privacy and security issues primarily involve the leakage of users' personal privacy and information leakage during communication. Due to these characteristics, eavesdroppers can acquire and eavesdrop on electromagnetic beams from any angle, thus threatening the secure transmission of private information. To address these challenges, information transmission security technologies in modern wireless communication networks are mainly divided into two categories. One is application-layer security transmission technology based on cryptography, which uses cryptographic encryption techniques to protect the confidentiality of data. The sender encrypts the transmitted signal using symmetric or asymmetric encryption algorithms, and the receiver must decrypt the signal using a key to recover it. The advantage of this encryption method is its excellent confidentiality and reliability. However, due to the rapid development of computer hardware and the emergence of supercomputers and quantum computers, cryptography-based information security technologies have encountered unprecedented challenges. The other type of information transmission security technology is physical-layer security transmission technology based on the principles of wireless communication. This technology utilizes the characteristics of the underlying protocol physical layer or network layer of the communication protocol network to achieve encrypted transmission of protected data. Physical layer security technologies employ various methods, such as channel coding, physical layer key generation, and authentication, to improve the signal-to-noise ratio (SNR) for legitimate users while suppressing the SNR of potential eavesdroppers, thereby achieving higher security performance. This technology leverages physical characteristics, making it difficult for eavesdroppers to crack the code through technical means, thus significantly improving the security and reliability of communication.

[0004] With the in-depth development of 5G technology, omnidirectional smart surfaces (IOS) are receiving extensive research as a new type of communication hardware. Omnidirectional smart surfaces not only possess the same functions as smart reflectors but also offer more significant advantages. The role of traditional intelligent reflectors (IRS) in communication systems has been thoroughly studied. Smart reflectors consist of controllable reflective elements, each of which can be programmed or algorithmically modified to change the phase angle and amplitude of the transmitted beam, thereby altering the wireless channel environment and improving the performance of the wireless system. Compared to traditional smart reflectors, IOS offers higher transmission efficiency and stability: IOS uses a different transmission method; traditional IRS primarily utilizes reflection to transmit signals, while IOS can utilize both transmission and reflection paths. IOS allows for more flexible deployment: traditional IRS requires a smart surface to be placed between the receiver and transmitter to adjust the signal amplitude and phase, while IOS can be deployed anywhere. IOS control is simpler: traditional IRS requires a central controller to adjust the phase and amplitude of the smart surface, while IOS technology uses distributed algorithms for adaptive control, resulting in lower control complexity and cost. Therefore, smart omnidirectional surfaces can not only improve the performance of overall wireless communication systems, but also offer advantages such as low cost and miniaturization, enabling large-scale deployment on building surfaces and indoor spaces. In particular, the ability of IOS to alter the wireless channel environment is expected to deliver outstanding performance in physical layer security.

[0005] First, physical layer security technology provides strong support for solving the problem of secure transmission, achieving security at the information theory level. Smart surfaces, as a novel communication technology, can be well integrated with physical layer security technology due to their ability to reconstruct the physical wireless environment. Currently, there are many research results on physical layer security assisted by traditional IRSs that only reflect signals. However, traditional IRSs only have the ability to reflect signals and can only cover users in the reflected area. IOS, as the latest smart surface, has significant advantages over IRSs, such as a wider coverage area and better communication performance. Summary of the Invention

[0006] To address the aforementioned problems in the prior art, this invention provides an omnidirectional smart surface-assisted physical layer secure transmission method. The technical problem to be solved by this invention is achieved through the following technical solution:

[0007] This invention provides an omnidirectional smart surface-assisted physical layer secure transmission method applied to a downlink MISO communication system. The downlink MISO communication system includes: a transmitting base station, a smart omnidirectional surface, a legitimate user, and an eavesdropper. The omnidirectional smart surface-assisted physical layer secure transmission method includes:

[0008] S1, the transmitting base station obtains channel information of legitimate users and eavesdroppers through channel estimation;

[0009] S2, the transmitting base station calculates the transmit beamforming vector according to the channel information, and transmits the user signal simultaneously via space division multiple access according to the transmit beamforming vector;

[0010] S3, the intelligent omnidirectional surface initializes the reflection and refraction beamforming vectors according to the channel information;

[0011] S4. Combining the transmit beamforming vector of the transmitting base station and the reflection and refraction beamforming vectors at the smart omnidirectional surface, an optimization problem is constructed to maximize the total security rate of the downlink MISO communication system.

[0012] S5, the optimization problem is transformed by constructing an upper limit approximation function to obtain an optimization problem with a convex function, and the transformed optimization problem is divided into two sub-optimization problems according to the sending stage;

[0013] S6. By constructing an approximate function, the two sub-optimization problems are iterated alternately, and the convergence is determined based on the result of each iteration. If convergence is achieved, the result of the last iteration is substituted into the optimization problem transformed in step 5 to obtain an approximate solution that maximizes the total security rate.

[0014] S7 securely transmits signals to legitimate users using an approximate solution that maximizes the sum of the confidential rates.

[0015] The beneficial effects of this invention are:

[0016] 1. This invention provides a physical layer secure transmission method assisted by an omnidirectional smart surface. By introducing a smart omnidirectional surface into the downlink MISO wireless communication network, since inactive users in each system have potential eavesdropping risks, the wireless transmission channel can be reconstructed, the legitimate reception channel can be enhanced, the eavesdropping channel can be weakened, and the system's security capacity can be improved by jointly designing the base station transmit beamforming vector and the smart omnidirectional surface passive beamforming vector.

[0017] 2. By calculating the lower bound of the security and rate of the system, this invention can guarantee the stability and security of transmission from an information theory perspective, as long as the transmitted information is transmitted at a rate not exceeding the security and rate.

[0018] 3. Through theoretical analysis, this invention reveals the impact of the number of IOS reflective elements, base station transmit power, base station transmit antenna, and eavesdropper location on physical layer security performance. Therefore, it proposes to jointly optimize the transmit beamforming vector and the IOS reflective beamforming vector to maximize system confidentiality and speed, and improve network security.

[0019] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0020] Figure 1 This is a model diagram of the downlink MISO network using omnidirectional smart surface assistance in this invention;

[0021] Figure 2 A flowchart of an omnidirectional smart surface-assisted physical layer secure transmission method is provided for this invention;

[0022] Figure 3 This is a flowchart illustrating the confidentiality rate alternating iterative optimization algorithm of the present invention;

[0023] Figure 4 This is a graph showing the convergence of the algorithm of this invention;

[0024] Figure 5 This is a performance comparison chart of the algorithm of this invention with other transmission schemes;

[0025] Figure 6 This is a graph showing the relationship between the security and rate of this invention and the number of transmit antennas and IOS reflective elements;

[0026] Figure 7 This is a diagram showing the relationship between the confidentiality and speed of the present invention and the location of the eavesdropper. Detailed Implementation

[0027] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0028] This invention provides an omnidirectional intelligent surface-assisted physical layer secure transmission method, which utilizes a downlink MISO wireless transmission network, such as... Figure 1 As shown, the system consists of one base station, two users, and two eavesdroppers. The base station is equipped with M antennas, while the users and eavesdroppers each have a single antenna. The base station provides communication services to both legitimate users simultaneously through active beamforming, with each user decoding their own information after receiving the mixed signal. The two eavesdroppers maliciously eavesdrop on legitimate communication and can decode the information of any one of the legitimate users. To ensure secure communication between the base station and users and reduce the threat posed by eavesdroppers, the system deploys an Intelligent Omni-Surface (IOS) with N reflective elements to minimize information leakage and achieve secure information transmission. The two users are located on opposite sides of the IOS. Since the signal experiences significant path loss after multiple reflections, which is negligible, this invention only considers first-order reflected signals. The IOS can communicate with the base station via a separate wireless link provided by the intelligent controller to coordinate and transmit information. Therefore, the base station can use the controller to achieve real-time control of the phase shift and amplitude of the IOS reflective elements.

[0029] Assume all channels are quasi-static block fading channels, and that reflection channels through adjacent components are independent. Due to obstruction, the direct transmission links from the base station to the user and the eavesdropper are blocked. The channels from IOS to the base station, user x, eavesdropper 1, and eavesdropper 2 are represented as follows: In the system studied in this invention, both large-scale and small-scale fading are considered. The channel from the base station to the IOS is a Ricean fading channel, and other channels are Rayleigh fading channels. In the wireless communication scenario studied in this invention, the eavesdroppers are inactive users among the legitimate users. That is, these eavesdroppers have also interacted with the base station via signaling, although they have not communicated. However, these inactive users have the potential to eavesdrop. Therefore, it is reasonable to assume that the base station possesses perfect CSI for both legitimate users and eavesdroppers.

[0030] like Figure 2 As shown, the present invention provides an omnidirectional smart surface-assisted physical layer secure transmission method, comprising:

[0031] S1, the transmitting base station obtains channel information of legitimate users and eavesdroppers through channel estimation;

[0032] In communication systems, CSI (Channel State Information) refers to information used to describe the channel state. Perfect CSI means that the receiver has complete knowledge of the channel's state information, including all information such as channel gain and phase. In channel modeling, perfect CSI includes the following:

[0033] Gain Information: Perfect CSI includes channel gain information, which is the attenuation of the transmitted signal in the channel. This is very important information because the signal attenuation determines the signal strength and reliability.

[0034] Phase information: In addition to gain information, perfect CSI also includes channel phase information. Phase information describes the phase changes of the transmitted signal as it travels through the channel, and is crucial for signal demodulation and recovery.

[0035] Time-varying information: In actual communication systems, the state of the channel is usually time-varying. Therefore, a perfect CSI should also include time-varying information, that is, information about how the channel state changes over time.

[0036] Error Information: A perfect CSI should also include error information, that is, information about errors that may occur during the channel state estimation process. Error information also has a significant impact on the accuracy and reliability of channel estimation.

[0037] The channel information from IOS to the base station, user χ, eavesdropper 1, and eavesdropper 2 is represented as follows:

[0038] S2, the transmitting base station calculates the transmit beamforming vector through channel information and transmits the user signal simultaneously through space division multiple access;

[0039] The transmission signal of the base station is

[0040]

[0041] Where, x χ This represents a signal with unit variance, given to the user χ. Represents the beamforming vector. Let M represent the vector space, and M represent the number of transmitting antennas. In the reflection zone of IOS, there are User 1 and Eavesdropper 1. The received signals at User 1 and Eavesdropper 1 are:

[0042]

[0043] in, Additive white Gaussian noise at the locations of User 1 and Eavesdropper 1 Indicates channel h ζ,S The transpose of h ζ,S Θ represents the channel from user ζ to IOS. r Let represent the reflection phase shift matrix of the IOS, s represent the base station transmitted signal, U1 represent user 1, and E1 represent eavesdropper 1; user 2 and eavesdropper 2 exist within the projection area of ​​the IOS, and the received signals of user 2 and eavesdropper 2 are as follows:

[0044]

[0045] Where, Θ t The transmission phase shift matrix of IOS is defined as follows: The reflection phase shift matrix and the transmission phase shift matrix are defined as follows: and These represent the nth element in the reflection phase shift matrix and the transmission phase shift matrix, respectively. U2 and E2 represent the phase shift adjustment ranges of the nth reflecting and transmitting elements, respectively. U2 represents user 2, and E2 represents eavesdropper 2. Due to the law of conservation of energy, ζ n ∈[0,1] represents the gain parameter of the nth element; The additive white Gaussian noise represents the noise at user 2 and eavesdropper 2.

[0046] Upon receiving superimposed signals from the base station, each legitimate user treats the signals of other users as interference and only decodes its own signal. Due to the full-space coverage of IOS, a new eavesdropping scenario arises: any eavesdropper can not only eavesdrop on user signals on the same side but also on user signals on the opposite side of IOS. Therefore, the reachable rate of a legitimate user χ is expressed as:

[0047]

[0048] Where, Θ r / t Represents the phase shift matrix of reflection or refraction. Indicates user The transmitted beamforming vector, σ 2 This represents the variance of Gaussian white noise;

[0049] The eavesdropping rate of the eavesdropper against user χ is expressed as:

[0050]

[0051] in, This indicates that the eavesdropper k is connected to the IOS channel. transpose,

[0052] Therefore, the confidentiality rate of any legitimate user can be defined as the difference between the achievable rate and the maximum eavesdropping rate of the two eavesdroppers:

[0053]

[0054] in, The confidentiality rate for legitimate users is not negative. This represents the rate at which eavesdropper 1 eavesdrops on user χ. This represents the rate at which eavesdropper 2 eavesdrops on user χ;

[0055] Based on equation (5), summing the security rates of all legitimate users, the total security rate of all users in the downlink MISO communication system, assuming no eavesdropping, is:

[0056]

[0057] in, This represents the reachable rate of user 1. This indicates the rate at which eavesdropper 1 eavesdrops on user 1. This indicates the rate at which eavesdropper 2 eavesdrops on user 1. This represents the reachable rate of user 2. This indicates the rate at which eavesdropper 1 eavesdrops on user 2. This indicates the rate at which eavesdropper 2 eavesdrops on user 2.

[0058] S3, the intelligent omnidirectional surface initializes the reflection and refraction beamforming vector Θ based on channel information. r / t ;

[0059] Where, Θ r / t Take the reflection phase shift matrix Θ in the reflection region. r In the transmission region, the transmission phase shift matrix Θ is taken.t ;

[0060] S4, combining the transmit beamforming vector of the transmitting base station and the reflection and refraction beamforming vectors at the smart omnidirectional surface, constructs an optimization problem that maximizes the total security rate of the downlink MISO communication system;

[0061] Based on the description in the previous section, the research objective of this invention is to jointly design the transmit beamforming vector w at the base station, under the constraints of base station transmit power and IOS continuous phase shift matrix. χ and the reflection phase shift matrix Θ at IOS r and transmission phase shift matrix Θ t To adjust the reflected and refracted beams, the system's security and speed are maximized. Therefore, the optimization problem is as follows:

[0062]

[0063] Among them, P max Let C represent the transmit power threshold of the base station. Constraint C1 is the total transmit power constraint at the base station, and constraints C2 and C3 are the amplitude and phase shift constraints of the IOS. Since the objective function is non-convex, the optimization problem P1 is difficult to solve.

[0064] For the sum of security rates expressed in equation (6), since R sec Variables Θ containing coupling r / t and w χ ,make Q χ =G H diag(h χ,S ), Then the total security rate R in equation (6) sec for:

[0065]

[0066] in, and It is a positive semi-definite matrix. This represents the transmit beam matrix of user 1. U represents the equivalent channel from the base station to user 1. r This represents the equivalent parameter matrix of IOS reflections. This represents the transmit beam matrix of user 2. U represents the equivalent channel from the base station to user 2. t Represents the IOS refraction equivalent parameter matrix;

[0067] To solve R sec The non-convexity problem of expressions can be solved by introducing slack variables. To R secThe boundary values ​​are represented.

[0068] The achievable rate for user χ can be handled as follows:

[0069]

[0070]

[0071]

[0072] Therefore, the lower bound of the achievable rate for user χ can be obtained:

[0073] R χ =l n,χ -l d,χ ≤R χ (13);

[0074] The upper bound of the eavesdropping rate for user χ can be defined as:

[0075]

[0076] in, express The rate of slack variable, This represents the slack variable in the denominator of the expression for the eavesdropping rate of user χ by eavesdropper k. The following constraints must be met:

[0077]

[0078]

[0079]

[0080] in, This represents the slack variable in the numerator of the expression for the eavesdropping rate of user χ by eavesdropper k.

[0081] The lower bound expression for the secrecy rate can be determined by using the lower bound expression for the achievable rate and the upper bound expression for the eavesdropping rate. Since equations (10) and (16) are non-convex, this invention constructs an upper bound approximation function B(x,y) for log2(x):

[0082]

[0083] The equality holds when y = x. Therefore, equations (10) and (16) can be transformed into:

[0084]

[0085] Wherein, B(μ) n,χ ,yχ ) represents log2(μ n,χ The convex approximation function of ). express The rate of slack variable, Represents log2 The convex approximation function, y χ and y k,χ B(μ) n,χ ,y χ )and Approximate parameters;

[0086] Therefore, the optimization problem P1 can be reformulated as the optimization problem P1.1:

[0087]

[0088] Among them, W χ Represents the transmitted beam matrix. U represents the maximum rate at which two eavesdroppers can eavesdrop on user χ. r,n,n Represents matrix U r The element in the nth row and nth column, U t,n,n Represents matrix U t The element in the nth row and nth column.

[0089] This invention utilizes a security rate alternating iterative optimization algorithm to divide the optimization problem P1.1 into two sub-optimization problems P2 and P3; the optimization algorithm is shown in Table 1.

[0090] Table 1. Algorithm for Alternating Iterative Optimization of Security Rate

[0091]

[0092] Active beamforming optimization

[0093] In this invention, the reflection or transmission phase shift matrix Θ of IOS is assumed. r / t It is a constant, i.e., u r / t and U r / t Assuming the variables are constant and removing irrelevant constraints, problem P1.1 can be simplified to:

[0094]

[0095] in, μn,χ For the slack variable in the denominator of the user's χ rate expression, Regarding the upper bound of the eavesdropping rate for User 1, The upper bound of the eavesdropping rate for User 2;

[0096] Optimization problem P2 can be solved by iteratively updating y. χ =μ n,χ and This can be solved by... The rank-one constraint can be ignored because rank(W)... χ )≤rank(U r / t ) = 1. Therefore, the convex optimization problem P2 can be effectively solved iteratively using the CVX toolbox.

[0097] IOS reflection and transmission phase optimization

[0098] In this invention, it is assumed that the active beam vector w k W is a constant. χ Assuming the variables are constant and removing irrelevant constraints, problem P1.1 can be simplified to:

[0099]

[0100] The rank-one constraint in constraint C12 is a non-convex constraint, which can be ignored by using SDR relaxation. To address the issue that ignoring the rank-one constraint leads to solutions that do not satisfy it, this invention uses Gaussian randomization to obtain approximate solutions. Therefore, optimization problem P3 can be solved using the CVX Convex Optimization Toolbox.

[0101] The proof is as follows:

[0102] The rank-one relaxation problem P2 is jointly convex with respect to the optimization variables and satisfies Slater's constraints. Therefore, the corresponding Lagrangian function for W is given by the following formula:

[0103]

[0104] Where, {λ1,λ2,X χ} represents the corresponding Lagrange multiplier, and ι represents the multiplier independent of W. χ The set. According to the KKT conditions:

[0105]

[0106] in Through some equivalent algebraic operations, it can be easily obtained that:

[0107] W * =CU t / r C H W * (25)

[0108] Therefore, it can be proven that:

[0109] rank(W * ) = rank(CU t / r C H W * )≤rank(U r / t)=1 (26)

[0110] Θ r / t The approximate solution is, if we do not omit the change to Subtract equations (13) and (14), and substitute the parameter values ​​obtained from each alternation into the formula for subtraction. Determine whether the result obtained after substitution is less than the preset threshold. If yes, then the alternation converges. If no, then continue the alternation iteration.

[0111] S7 securely transmits signals to legitimate users using an approximate solution that maximizes the sum of the confidential rates.

[0112] The performance of the present invention will be further described below in conjunction with simulation experiments.

[0113] Figure 3 The relationship between security and rate of return and the number of iterations is presented in the figure. It can be seen from the figure that when the number of base station transmit antennas and IOS reflective elements is the same, the security and rate of return increase with the number of iterations and then tend to converge. The security and rate of return increase with increasing M and N, while the convergence speed of the algorithm decreases with increasing N. This is because as M or N increases, more spatial degrees of freedom are available, resulting in better optimization performance. However, as N increases, the complexity of the algorithm also increases significantly, leading to slower convergence.

[0114] Figure 4 This paper demonstrates that as the upper limit of base station transmit power increases, both security and data rate also increase. Three sets of comparison results clearly show that the IOS alternating optimization scheme performs best, followed by the traditional RIS alternating optimization scheme, while the IOS random phase scheme performs worst. The MRT and ZF schemes, as mature transmit beamforming schemes in MIMO technology, have poor algorithm performance because they do not jointly consider IOS reflection beamforming or the eavesdropper channel situation. Simulation results show that, under any scheme, modifying the number of reflection elements N, a larger N can still achieve better optimization results. Interestingly, simulation results show that the traditional RIS-assisted scheme combined with a larger number of 20 reflection elements performs better than the IOS-assisted scheme combined with a smaller number of 10 reflection elements, indicating that increasing the spatial degrees of freedom brought by increasing N has a greater impact on communication.

[0115] Figure 5The relationship between system security and data rate and the maximum transmit power of the base station is presented under different base station antenna numbers M and IOS reflection / transmission elements N. The figure shows that with the same N, a larger M yields better optimization results. Furthermore, with the same M, the conclusion from previous simulations remains the same: a larger N brings better optimization gain. Additionally, considering the baseline parameter setting of 4M10N (representing 4 transmit antennas and 10 IOS elements), we compared M and N by doubling them, i.e., comparing the 8M10N and 4M20N schemes. We found that doubling N significantly outperformed increasing M, and even the 4M15N scheme outperformed the 8M10N scheme. This indicates that increasing N is more effective than increasing M, but the gain from increasing M is not negligible.

[0116] Figure 6 The relationship between base station transmit power and the distance between the eavesdropper and the IOS is presented. When the distance between the eavesdropper and the IOS is less than that of all legitimate users, the algorithm has no feasible region and no solution. This is because this invention considers the worst-case scenario for the eavesdropper, where there is no interference from other users, only noise interference. However, users receiving signals need to consider interference from other users' signals. Therefore, if the eavesdropper is too close to the IOS, the eavesdropping rate cannot be guaranteed to be less than a threshold. As the distance between the eavesdropper and the IOS increases, the base station transmit power also decreases and gradually converges to a certain value. This is because as the distance increases, the eavesdropper's eavesdropping rate constraint is no longer the main factor affecting the transmit power; the minimum information rate at the user's location becomes the main constraint, and the transmit power therefore tends to converge.

[0117] This invention provides an omnidirectional intelligent surface-assisted physical layer secure transmission method. By incorporating IOS-assisted MISO wireless communication and jointly optimizing the active beamforming vector at the base station and the passive beamforming vector at the IOS, it maximizes system security and data rate. To solve the proposed non-convex optimization problem, this invention proposes an alternating optimization algorithm. For the non-convex objective function, an approximation function is constructed to approximate it into a convex form. The active and passive beamforming vectors are alternately optimized. Through simulation analysis, this invention not only improves the received signal strength of legitimate users but also suppresses the received signal strength of eavesdroppers, significantly improving the system's security and data rate performance.

[0118] Although this application has been described herein in conjunction with various embodiments, those skilled in the art will understand and implement other variations of the disclosed embodiments by reviewing the accompanying drawings, the disclosure, and the appended claims in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality.

[0119] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A method for secure physical layer transport assisted by an omnidirectional intelligent surface, characterized in that, Applied to a downlink MISO communication system, the downlink MISO communication system includes: a transmitting base station, a smart omnidirectional surface, a legitimate user, and an eavesdropper; the omnidirectional smart surface-assisted physical layer secure transmission method includes: S1, the transmitting base station obtains channel information of legitimate users and eavesdroppers through channel estimation; S2, the transmitting base station calculates the transmit beamforming vector according to the channel information, and transmits the user signal simultaneously via space division multiple access according to the transmit beamforming vector; S3, the intelligent omnidirectional surface initializes the reflection and refraction beamforming vectors according to the channel information; S4, combining the transmit beamforming vector of the transmitting base station and the reflection and refraction beamforming vectors at the intelligent omnidirectional surface, constructs an optimization problem that maximizes the total security rate of the downlink MISO communication system; S4 includes: By jointly designing the transmit beamforming vector at the base station Reflection phase shift matrix at IOS Transmission phase shift matrix To adjust the reflected and refracted beams and maximize the system's security and speed, an optimization problem is constructed: （8）； in, This represents the sum of the security rates. This represents the transmit power threshold of the base station. Constraint C1 is the total transmit power constraint at the base station, constraint C2 is the amplitude constraint of the IOS, and constraint C3 is the phase shift constraint of the IOS. Represents the active beam vector. Indicates user , , Representing a vector space, Indicates the number of transmitting antennas. This represents the gain parameter of the nth element. These represent the phase shift adjustment ranges for the nth reflecting and transmitting elements, respectively. S5, the optimization problem is transformed by constructing an upper limit approximation function to obtain an optimization problem with a convex function, and the transformed optimization problem is divided into two sub-optimization problems according to the sending stage; S6. By constructing an approximate function, the two sub-optimization problems are iterated alternately, and the convergence is determined based on the result of each iteration. If convergence is achieved, the result of the last iteration is substituted into the optimization problem transformed in step 5 to obtain an approximate solution that maximizes the total security rate. S7 securely transmits signals to legitimate users using an approximate solution that maximizes the sum of the confidential rates.

2. The omnidirectional intelligent surface-assisted physical layer secure transmission method according to claim 1, characterized in that, The transmission signal of the base station in S2 is: （1）； in, Indicate to the user A signal with unit variance is present in the reflection zone of IOS, where User 1 and Eavesdropper 1 are located. The received signals at User 1 and Eavesdropper 1 are as follows: （2）； in, This indicates the channel information from the IOS to the base station. Additive white Gaussian noise at the locations of User 1 and Eavesdropper 1 Indicates channel The conjugate transpose of . Indicates user Channel to iOS This represents the reflection phase shift matrix of IOS. This indicates that the base station is transmitting a signal. Indicates user 1, Indicates eavesdropper 1; Within the projection area of ​​the iOS device, there are User 2 and Eavesdropper 2. The received signals of User 2 and Eavesdropper 2 are as follows: （3）； The reflection phase shift and transmission phase shift matrices are defined as follows: and , , These represent the nth element in the reflection phase shift matrix and the transmission phase shift matrix, respectively. Indicates user 2, This refers to eavesdropper 2, due to the law of conservation of energy. This represents the gain parameter of the nth element; The additive white Gaussian noise represents the noise at user 2 and eavesdropper 2.

3. The omnidirectional intelligent surface-assisted physical layer secure transmission method according to claim 2, characterized in that, The reachable rate for legitimate users is represented as: （4）； in, Represents the phase shift matrix of reflection or refraction. Indicates user The transmitted beamforming vector, This represents the variance of Gaussian white noise. Indicates user The conjugate transpose of the channel information to IOS; The eavesdropper's eavesdropping rate is expressed as: （5）； in, , This indicates that the eavesdropper k is connected to the IOS channel. The conjugate transpose of . The confidentiality rate for any legitimate user is the difference between the achievable rate and the maximum eavesdropping rate of the two eavesdroppers, expressed as: （6）； in, For legitimate users, the confidentiality rate is not negative. This indicates that eavesdropper 1 was eavesdropping on users. rate, This indicates that the eavesdropper 2 was eavesdropping on users. The rate; Based on equation (5), summing the security rates of legitimate users, we obtain the total security rate of all users in the downlink MISO communication system, assuming no eavesdropping: （7）； in, This represents the reachable rate of user 1. This indicates the rate at which eavesdropper 1 eavesdrops on user 1. This indicates the rate at which eavesdropper 2 eavesdrops on user 1. This represents the reachable rate of user 2. This indicates the rate at which eavesdropper 1 eavesdrops on user 2. This indicates the rate at which eavesdropper 2 eavesdrops on user 2.

4. The omnidirectional intelligent surface-assisted physical layer secure transmission method according to claim 3, characterized in that, In S5, the optimization problem is transformed by constructing an upper bound approximation function, resulting in optimization problems exhibiting convex functions, including: Regarding the total security rate, due to Variables containing coupling and ,make , , , Then the sum of the security rates in equation (6) for: （9）； in, and It is a positive semi-definite matrix. This represents the transmit beam matrix of user 1. This represents the equivalent channel from the base station to user 1. Represents the IOS reflection equivalent parameter matrix. This represents the transmit beam matrix of user 2. This represents the equivalent channel from the base station to user 2. Represents the IOS refraction equivalent parameter matrix; For users The achievable rate can be processed as follows: （10）； （11）； （12）； in, superscript Indicates targeting users The slack variable in the numerator of the rate expression. superscript express The approximation of the rate form, Indicates targeting users The slack variable in the denominator of the rate expression; can be obtained from the user. Lower bound of achievable rate: （13）； For users The upper bound of the eavesdropping rate can be defined as: （14）； in, express The rate of slack variable, This indicates that the eavesdropper k is targeting the user. The slack variable in the denominator of the eavesdropping rate expression. The following constraints must be met: （15）； （16）； （17）； in, This indicates that the eavesdropper k is targeting the user. The slack variable in the numerator of the eavesdropping rate expression; The lower bound expression for the secure rate can be determined by using the lower bound expression for the achievable rate and the upper bound expression for the eavesdropping rate. Since equations (10) and (16) are non-convex, construct a upper limit approximation function : （17）； The equality relationship is in When this holds true, equations (10) and (16) can be transformed into: （18）； in, express The convex approximation function, express The rate of slack variable, express The convex approximation function, and They are respectively and Approximate parameters; Therefore, the optimization problem P1 is transformed into a convex function form, expressed as: （20）； in, Represents the transmitted beam matrix. This indicates that two eavesdroppers are targeting the user. The maximum eavesdropping rate, Representation matrix The element in the nth row and nth column, Representation matrix The element in the nth row and nth column.

5. The omnidirectional intelligent surface-assisted physical layer secure transmission method according to claim 4, characterized in that, In S5, the transformed optimization problem is divided into two sub-optimization problems according to the sending stage: The alternating optimization algorithm is used to transform the optimization problem P1.1 into two sub-problems P2 and P3; Among them, the optimization sub-problem P1 is obtained by optimizing the transmitted beamforming vector at the transmitting base station, and the reflection or transmission phase shift matrix of IOS in the optimization sub-problem P1 is... The constant is used; the optimization subproblem P2 is obtained by optimizing the IOS reflection and refraction beamforming vectors, and the optimization subproblem P1 contains the active beam vector. It is a constant.

6. The omnidirectional intelligent surface-assisted physical layer secure transmission method according to claim 5, characterized in that, Assuming the reflection or transmission phase shift matrix of IOS Assuming the variables are constant and removing irrelevant constraints, the optimization problem P1.1 simplifies to: （21） in, For users The slack variable in the denominator of the rate expression. Regarding the upper bound of the eavesdropping rate for User 1, The upper bound of the eavesdropping rate for User 2; Optimization problem P2 can be solved through iterative updates. and To solve, since Therefore, the rank-one constraint is ignored; Assuming active beam vector Assuming the variables are constant and removing irrelevant constraints, the optimization problem P1.1 simplifies to: （26）。 7. The omnidirectional intelligent surface-assisted physical layer secure transmission method according to claim 6, characterized in that, S6 include: (a) The confidentiality rate of the two sub-optimization problems is processed into a convex function form by constructing an approximate function; (b) The optimization problem P2 of convex function formation is solved using the SDR relaxation algorithm and CVX to obtain the source beam vector. ; (c) The obtained source beam vector Introducing the optimization problem P3, we obtain the reflection phase shift matrix or transmission phase shift matrix of IOS. ; (d) Alternate between (b) and (c) to obtain the reflection or transmission phase shift matrix containing IOS in each iteration. Source beam vector The parameter results are used to determine whether convergence has occurred. If convergence has occurred, the result of the last iteration is substituted into the transformed optimization problem in step 5 to obtain an approximate solution that maximizes the total security rate. If convergence has not occurred, (b) and (c) are continued alternately until convergence is achieved.

Images