Intelligent transmission surface and intelligent reflection surface combined assisted robust secure transmission method

Abstract

The invention discloses an intelligent transmission surface and intelligent reflection surface combined assisted robust secure transmission method, which considers the condition that a direct channel between a BS and a Bob is shielded by an obstacle, utilizes a bounded error model to perform modeling on uncertainty of an eavesdropping reflection channel, and aims at weighting and safety rate maximization. A multi-user power distribution coefficient, ITS beam forming and IRS phase shifting are jointly optimized by taking base station transmitting power constraints, ITS transmitting element constraints and unit modulus IRS phase shifting constraints into consideration, and the unit modulus constraints of the IRS phase shifting are processed by using continuous convex approximation, alternating optimization and Cauchy-Schwartz inequality and adopting a penalty method. According to the robust beam forming method, signal transmission under the non-ideal condition is considered, and the robust beam forming method is suitable for an ITS and IRS combined auxiliary system.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communications, and in particular relates to a robust and secure transmission method based on the joint assistance of an intelligent transmission surface and an intelligent reflection surface. Background Art

[0002] Reconfigurable smart surfaces (RIS) are considered a promising technology for future wireless communication systems, as they have been shown to significantly improve the spectral and energy efficiency of networks. By controlling the microstructure of RIS units, their electromagnetic (EM) response can be changed to precisely adjust the amplitude, phase, polarization, and frequency of the incident electromagnetic wave. This allows the signal to be precisely reflected or transmitted, thereby achieving active intervention in the propagation of wireless signals. In addition to being deployed on the channel side for precise reflection or transmission, RIS can also be used to build transceivers to achieve high-gain beams with low power consumption through intelligent control of electromagnetic waves.

[0003] In order to further improve the network performance, people have studied the joint assistance system of intelligent transmission surface (ITS) and intelligent reflection surface (IRS). The multiple input multiple output (MIMO) network assisted by the intelligent transmission surface and the intelligent reflection surface has achieved the maximization of weighted sum security rate and energy efficiency by alternately optimizing the beamforming of the base station (BS) and the phase shift of the intelligent transmission surface and the intelligent reflection surface. Based on this, the prior art discloses a cell-free system assisted by ITS and IRS, in which the system total rate is maximized. In terms of confidential communication, the performance of ITS and IRS assisted systems in Internet of Things (IoT) networks and satellite communications has been studied respectively, and two low-complexity algorithms have been proposed to achieve the maximum weighted sum security rate (WSSR) and security energy efficiency (SEE).

[0004] However, current research is limited to the scenario of ideal channel state information (CSI). In practical applications, it is challenging to obtain accurate CSI for IRS-assisted systems because IRS lacks a radio frequency (RF) chain for baseband processing. To cope with the inevitable channel uncertainty, some studies have proposed robust beamforming schemes to achieve the minimum transmit power of the base station in the case of imperfect reflection CSI and imperfect cascaded CSI. For confidential communication, the prior art considers the statistical CSI error model of the cascaded eavesdropping channel in the IRS-assisted security network and uses the Bernstein-type inequality to deal with the confidentiality interruption probability constraint. In addition, the total rate maximization problem under the channel capacity constraint of the eavesdropper is studied, and the generalized S-procedure is used to solve the infinite linear matrix inequality (LMI) constraint introduced by the norm-bounded CSI error model of the IRS-eavesdropper (Eve) channel. For the scenario of multiple Eves, the prior art studies the problem of maximizing the security energy efficiency (SEE) in the case of imperfect CSI of all Eve-related channels. However, the existing robust algorithms are not applicable to the joint ITS and IRS-assisted system.

[0005] Therefore, a robust algorithm suitable for the joint ITS and IRS assisted system is needed. Summary of the invention

[0006] In view of this, in order to address the deficiencies of the above-mentioned prior art, the object of the present invention is to provide a robust and secure transmission method based on the joint assistance of an intelligent transmission surface (ITS) and an intelligent reflection surface (IRS), and to solve the maximized weighted and secure rate under the constraints of base station transmission power, ITS beam and IRS phase shift. The robust algorithm is suitable for a joint ITS and IRS assisted system.

[0007] To achieve the above purpose, the technical solution adopted by the present invention is: A robust secure transmission method based on the joint assistance of an intelligent transmission surface (ITS) and an intelligent reflection surface (IRS), wherein the confidential communication network relied on by the method includes a base station (BS) based on the ITS, an IRS, L single-antenna Bobs and L single-antenna passive eavesdroppers (Eves); wherein the ITS includes N transmitting elements and the IRS includes M reflecting elements; the BS uses the ITS transmitter to simultaneously send private information to all Bobs in the same frequency band under the monitoring of Eves; the method includes the following steps: S1. Build a signal transmission model to determine the ITS beamforming vector, IRS phase shift vector, and Bob / Eve’s received signal. S2, assuming that the reflection channel is non-ideal, the bounded error model is used to model the uncertainty of the eavesdropping reflection (IRS-Eves) channel to determine the reflection channel and the security rate; S3, with the goal of maximizing the weighted and secure rate WSSR, a joint robust design of multi-user power allocation coefficients, ITS beamforming and IRS phase shift is performed to determine the optimization problem P1; S4, using the successive convex approximation technique SCA to transform the objective function of the optimization problem P1 into an approximate concave form of the problem P2, and obtain the optimization variables in the objective function; S5. Use the alternating optimization AO method to decouple problem P2, and alternately optimize the optimization variables to find the optimal solution to the objective function of problem P2.

[0008] Furthermore, the model is multi-user multi-input single-output.

[0009] Further, in S1, let represents the beamforming vector of ITS, where β n ∈[0,1] and α n ∈[0,2π) represent the amplitude and phase responses of the nth transmission element of the ITS, respectively.

[0010] Furthermore, in S2, it is assumed that the reflection channel is imperfect, that is: in, and They represent the links from BS to IRS and from IRS to the lth Bob / Eve (l=l…L) respectively; e b,l and e e,l denote the estimated value and estimated error of the legal channel coefficient and the eavesdropping channel coefficient respectively; according to the criterion constraint error model, we have ||e b,l ||≤ξ b,l and ||e e,l ||≤ξ e,l ,l=1...L,where ξ b,l and e,l is the upper limit of the estimation error; Assuming that each Eve tries to eavesdrop on the Bob closest to her, the security rate of the lth Bob / Eve is: a l ≥0 is every s l The power allocation coefficient of θ = diag(θ) represents the additive white Gaussian noise (AWGN) at the lth Bob; Therefore, WSSRR s is defined as: Among them, η lRepresents the weight coefficient of the lth Bob.

[0011] Furthermore, in S3, with the goal of maximizing the weighted and safe rate, and considering the base station transmit power allocation constraints, ITS beam constraints and IRS phase shift constraints, a mathematical model of the dimensional coordinates and path loss index parameters is established to determine the maximization system weighted and safe rate WSSR optimization problem P1 as follows: Among them, P s is the transmit power of the BS, and constraints C2 and C3 represent the ITS transmit element constraint and the unit modulus constraint of the IRS phase shift, respectively.

[0012] Furthermore, in S5, the problem P2 is decoupled into three sub-problems by using alternating optimization, and the multi-user power allocation coefficient is alternately solved. The ITS beamforming vector w and the IRS phase shift vector θ are optimized until the convergence accuracy is met or the maximum number of iterations is reached.

[0013] Furthermore, in S5, the multi-user power allocation coefficient is optimized alternately. The ITS beamforming vector w and the IRS phase shift vector θ specifically include the following steps: S51, given w and θ, optimize by Taylor expansion and Cauchy-Schwartz inequality S52, given and θ, optimize w through SCA; S53, given and w, optimize θ by SCA and penalty method; S54, alternately iterate the above steps S51-S53 to find the optimal solution

[0014] The beneficial effects of the present invention are: The present invention provides a robust beamforming design method, which aims to maximize the system WSSR, considers the base station transmission power constraint, ITS transmission element constraint and unit modulus IRS phase shift constraint, and jointly optimizes the multi-user power allocation coefficient, ITS beamforming and IRS phase shift; By adopting continuous convex approximation (SCA), alternating optimization (AO) and Cauchy-Schwartz inequality, the present invention provides an optimization method for effectively processing the unit modulus constraint of the phase shift of the intelligent reflecting surface (IRS), and further optimizes it in combination with the penalty method; this method not only effectively solves the problems of non-convexity and computational complexity in the optimization problem, but also can maintain the stability of the system when the channel state information (CSI) is imperfect. Verified by numerical simulation results, this scheme can significantly improve the energy efficiency and safety performance of the system, especially in the application of multi-user MISO networks with extremely high robustness and reliability. In summary, the present invention provides a feasible solution for maximizing safety and energy efficiency in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 is a system model of the method of the present invention; Figure 2 is a flow chart of the method of the present invention; Figure 3 A flow chart of solving the optimization problem by the method of the present invention; Figure 4 is a curve of the convergence behavior change of the AO algorithm proposed in the embodiment of the present invention; Figure 5 is a relationship curve between WSSR and the number of IRS units in an embodiment of the present invention; Figure 6 is a relationship curve between WSSR and the number of ITS units in an embodiment of the present invention; Figure 7 : is the relationship curve between WSSR and transmission power in the embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following specific examples are given to further clearly, completely and in detail illustrate the technical solution of the present invention. This example is the best example based on the technical solution of the present invention, but the protection scope of the present invention is not limited to the following examples.

[0018] Example 1 A robust secure transmission method based on the joint assistance of an intelligent transmission surface (ITS) and an intelligent reflection surface (IRS), wherein the confidential communication network relied on by the method includes a base station (BS) based on ITS, an IRS, L single-antenna Bobs and L single-antenna passive eavesdroppers (Eves); wherein the ITS contains N transmitting elements and the IRS contains M reflecting elements; the BS uses the ITS transmitter to simultaneously send private information to all Bobs in the same frequency band under the monitoring of Eves.

[0019] The method comprises the following steps: Step 1: Build a signal transmission model to determine the ITS beamforming vector, IRS phase shift vector, and Bob / Eve's received signal; The model is a multi-user multi-input single-output model. represents the beamforming vector of ITS, where β n ∈[0,1] and α n ∈[0,2π) represent the amplitude and phase responses of the nth transmission element of the ITS, respectively; let represents the IRS phase shift vector, where represents the phase response of the mth reflection unit in the IRS; then the received signal of the lth Bob / Eve is: Among them, s l Represents the confidential information sent to the lth Bob, each s l are all assumed to be iid complex Gaussian variables, that is a l ≥0 is every s l The power allocation coefficient of k∈{b,e} represents the additive white Gaussian noise (AWGN) at the lth Bob / Eve respectively.

[0020] Step 2: Assuming that the reflection channel is non-ideal, the uncertainty of the eavesdropping reflection (IRS-Eves) channel is modeled using a bounded error model to determine the reflection channel and the security rate; Assume that the reflection channel is imperfect, that is: in, and They represent the links from BS to IRS and from IRS to the lth Bob / Eve (l=l…L) respectively; e b,l and e e,ldenote the estimated value and estimated error of the legal channel coefficient and the eavesdropping channel coefficient respectively; according to the criterion constraint error model, we have ||e b,l ||≤ξ b,l and ||e e,l ||≤ξ e,l ,l=1...L,where ξ b,l and e,l is the upper limit of the estimation error; Assuming that each Eve tries to eavesdrop on the Bob closest to her, the security rate of the lth Bob / Eve can be given by the following formula: Therefore, WSSRR s is defined as: Among them, η l Represents the weight coefficient of the lth Bob.

[0021] Step 3: To maximize the weighted and secure rate WSSR, perform joint robust design of multi-user power allocation coefficients, ITS beamforming, and IRS phase shift to determine the optimization problem P1: P1: Among them, P s is the transmission power of the BS; constraints C2 and C3 represent the unit modulus constraints of the ITS transmitting element and the IRS phase shift, respectively.

[0022] Step 4: Due to the non-concavity of the objective function and the deep coupling of the optimization variables, problem P1 is difficult to solve directly. The successive convex approximation technique SCA is used to transform the objective function of the optimization problem P1 into an approximately concave problem P2 to obtain the optimization variables in the objective function. Using SCA technology, problem P1 is transformed into the following form P2: C1-C3 in is the slack variable.

[0023] Step 5: Use the alternating optimization AO method to decouple problem P2 and design the multi-user power allocation coefficient The ITS beamforming vector w and the IRS phase shift vector θ are alternately optimized to solve the optimal solution of the objective function of problem P2; the following steps are included: S51. Optimize given w and θ The corresponding sub-questions are as follows: P2.1: stC4-C9,C1 Obviously, constraints C5, C6, C7, C8 and C9 are It is non-convex, and the constraint C5 can be converted into the following convex form using the Taylor expansion formula: C10: Note that for complex numbers p and q, the following inequality holds: ‖p|-|q‖≤|p+q|≤|p|+|q|; Combining the equation in step 2 with the above inequality using the Cauchy-Schwartz inequality, The upper and lower bounds of can be derived as follows: Where v∈{b,e}, it should be noted that in the derivation of equation When , assuming This is correct in most cases, since the estimation error is usually much smaller than the channel estimate; Substituting the constraints C7 and C8 and combining them with the Taylor expansion formula, we can get: C12: Put the equation Substituting constraints C6 and C9, we can obtain: C13: C14: Notice is a non-negative value, so problem P2.1 can be expressed in the following form: stC10,C11,C12,C13,C14,C4,C1 This problem is a linear programming (LP) problem and can be solved efficiently using CVX.

[0024] S52, given To optimize w under the condition of and θ, the corresponding sub-problems are as follows: P2.2: stC10,C11,C12,C13,C14,C4,C2 The constraints C13 and C14 are non-convex with respect to w. , equation C13 can be transformed into: in The convex form difference of the left side of the above inequality makes it still non-convex at w; Continuing to use the SCA technique, the above inequality can be approximated as: C15: Among them, w( t ) represents the solution of the tth iteration; similarly, constraint C14 can be transformed into: C16: After the above transformation, problem P2.2 is converted into the following form: stC10,C11,C12,C15,C16,C4,C2 The problem is a quadratically constrained quadratic programming (QCQP) problem in units of w, which can be solved using CVX.

[0025] S53, given To optimize θ under the condition of and w, the corresponding sub-problems are as follows: P2.3: stC10,C11,C12,C13,C14,C4,C3 Among them, C13, C14 and C3 are non-convex with respect to θ; Since ΘFw = diag(Fw)θ, constraint (14) can be rewritten as: in, Similar to S52, the convex difference on the left side of the above equation can be converted into a convex difference using SCA technology, thus approximating the equation as: C17: Similarly, constraint C14 can be transformed into: C18: At this point, question P2.3 can be organized as follows: stC10,C11,C12,C17,C18,C4,C3 Next, we use the penalty method to deal with the non-convex constraint C: First, the expression |θ m | 2 =1,m=1,...,M is equivalent to 1≤|θ m | 2 ≤1,m=1,...,M. Its non-convex part is 1≤|θ m | 2 ,m=1,...,M can be approximated by Taylor expansion formula as m=1,...,M; Then, introduce the slack variable p = [p 1 ,...,p 2M ] T Based on the penalty function method, the problem is rewritten as follows: stC10,C11,C12,C17,C18,C4, θ m | 2 ≤1+p m+M ,m=1,...,M, where λ (t) is the penalty factor. This problem is convex and can be solved directly using CVX.

[0026] The simulation results are as follows: like Figure 1 The figure shows the three-dimensional coordinates (in meters) of the simulation scene. The cooperative jammer and IRS are located at (0,0,10) and (0,100,10) respectively. Bobs are randomly located in a circular area with a radius of 10 meters centered at (20,100,1.)5. Each Eve is randomly scattered in a circular area with a radius of 5 meters centered at each Bob.

[0027] In this section, the robust performance of the proposed method is evaluated through MATLAB numerical simulations.

[0028] The three-dimensional coordinates of BS and IRS are (0,0,10) meters and (0,30,10) meters respectively; Bobs and Eves are randomly deployed on a circle with a radius of 3 meters, and the centers are (60,10,0) and (75,-5,0) respectively. Large-scale fading is modeled as PL=PL 0 -10αlog 10 (d / d 0 ), where d represents the distance between nodes, d 0represents the reference distance, which is set to 1 meter. Small-scale fading is modeled as Rician fading. Small-scale fading is modeled as Rician fading. Both ITS and IRS are modeled as uniform planar arrays. The Rician factor κ for all links h are all set to 5, and the path loss exponent of all links is set to 2.2. The normalized upper limit of the error between the legitimate channel and the eavesdropping channel is defined as where k∈{b,e}.

[0029] Unless otherwise specified, other parameters are configured as N=5, M=10, P s =15dBm, And L=4.

[0030] Figure 4-Figure 7 The performance evaluation of the proposed system design is given; Figure 4 The convergence behavior change curve of the proposed AO algorithm shows that the algorithm has good convergence performance. When the number of iterations increases, the weighted sum confidentiality rate (WSSR) of the system gradually increases and reaches a stable state within 25 iterations. With the increase of transmit power, the WSSR of all schemes increases, but the proposed algorithm has a significant performance improvement compared with the non-robust scheme, and is only slightly lower than the ideal CSI scheme, which verifies the robustness of the algorithm. Figure 5 is the relationship curve between WSSR and the number of IRS units, Figure 6 The relationship curve between WSSR and the number of ITS units further proves that increasing the number of IRS and ITS units can effectively improve WSSR, emphasizing the importance of these units in the optimal design. Figure 7 This is the relationship curve between WSSR and transmit power, which proves that the scheme proposed in the present invention has the potential in performance improvement and robustness in complex communication environments.

[0031] Through the above numerical simulation evaluation, the effectiveness of the scheme proposed in the present invention in improving the security rate and energy efficiency performance of the communication method is further verified. The established optimization model studies the joint auxiliary scheme of ITS and IRS under non-ideal conditions and has achieved remarkable results. It has good practicality and feasibility while improving system performance.

[0032] The above shows and describes the main features, basic principles and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements according to actual conditions, and these changes and improvements fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A robust secure transmission method jointly assisted by a smart transmission surface and a smart reflection surface, the method relies on a confidential communication network including an ITS-based base station BS, an IRS, L single-antenna Bobs and L single-antenna passive eavesdroppers Eves; The ITS includes N transmitting elements and the IRS includes M reflecting elements. The BS uses the ITS transmitter to send private information to all Bobs simultaneously in the same frequency band under the monitoring of Eves. The characteristics are: The method comprises the following steps: S1. Build a signal transmission model to determine the ITS beamforming vector, IRS phase shift vector, and Bob / Eve’s received signal. S2, assuming that the reflection channel is non-ideal, use the bounded error model to model the uncertainty of the eavesdropping reflection channel and determine the reflection channel and the security rate; S3, with the goal of maximizing the weighted and secure rate WSSR, a joint robust design of multi-user power allocation coefficients, ITS beamforming and IRS phase shift is performed to determine the optimization problem P1; S4, using the successive convex approximation technique SCA to transform the objective function of the optimization problem P1 into an approximate concave form of the problem P2, and obtain the optimization variables in the objective function; S5. Use the alternating optimization AO method to decouple problem P2, and alternately optimize the optimization variables to find the optimal solution to the objective function of problem P2.

2. The robust and secure transmission method assisted by a smart transmission surface and a smart reflection surface according to claim 1, characterized in that: The secure communication network model is multi-user multi-input single-output.

3. The robust and secure transmission method assisted by a smart transmission surface and a smart reflection surface according to claim 1, characterized in that: In S1, let represents the beamforming vector of ITS, where β n ∈[0,1] and α n ∈[0,2π) represent the amplitude and phase responses of the nth transmission element of the ITS, respectively.

4. The robust and secure transmission method assisted by a smart transmission surface and a smart reflection surface according to claim 3, characterized in that: In S2, it is assumed that the reflection channel is imperfect, that is: in, and They represent the links from BS to IRS and from IRS to the lth Bob / Eve (l=l…L) respectively; e b,l and e e,l denote the estimated value and estimated error of the legal channel coefficient and the eavesdropping channel coefficient respectively; according to the criterion constraint error model, we have ||e b,l ||≤ξ b,l and ||e e,l ||≤ξ e,l ,l=1...L,where ξ b,l and e,l is the upper limit of the estimation error; Assuming that each Eve tries to eavesdrop on the Bob closest to her, the security rate of the lth Bob / Eve is: a l ≥0 is every s l The power allocation coefficient of θ = diag(θ) represents the additive white Gaussian noise at the lth Bob; Therefore, WSSRR s is defined as: Among them, η l Represents the weight coefficient of the lth Bob.

5. The robust and secure transmission method assisted by a smart transmission surface and a smart reflection surface according to claim 1, characterized in that: In S3, the goal is to maximize the weighted and safe rate, and considering the base station transmission power allocation constraints, ITS beam constraints and IRS phase shift constraints, a mathematical model of the dimensional coordinates and path loss index parameters is established to determine the maximization system weighted and safe rate WSSR optimization problem P1 as follows: Among them, P s is the transmit power of the BS, and constraints C2 and C3 represent the ITS transmit element constraint and the unit modulus constraint of the IRS phase shift, respectively.

6. The robust and secure transmission method assisted by a smart transmission surface and a smart reflection surface according to claim 1, characterized in that: In S5, the problem P2 is decoupled into three sub-problems by using alternating optimization, and the multi-user power allocation coefficient is alternately solved. The ITS beamforming vector w and the IRS phase shift vector θ are optimized until the convergence accuracy is met or the maximum number of iterations is reached.

7. The robust and secure transmission method assisted by a smart transmission surface and a smart reflection surface according to claim 1, characterized in that: In S5, the multi-user power allocation coefficient is optimized alternately. The ITS beamforming vector w and the IRS phase shift vector θ specifically include the following steps: S51, given w and θ, optimize by Taylor expansion and Cauchy-Schwartz inequality S52, given and θ, optimize w through SCA; S53, given and w, optimize θ by SCA and penalty method; S54, alternately iterate the above steps S51-S53 to find the optimal solution

Images