Robust beam forming design method of intelligent reflector-assisted downlink MISO system

Abstract

The invention provides a robust beam forming design method for an intelligent reflector-assisted downlink MISO (Multiple Input Single Output) system, which comprises the following steps: establishing the intelligent reflector-assisted downlink MISO system which comprises a base station, a user, an intelligent reflector, K active eavesdroppers and an unknown number of passive eavesdroppers; constructing a user signal-to-noise ratio model and an eavesdropper signal-to-noise ratio model according to the intelligent reflector-assisted downlink MISO system; according to the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model, establishing an optimization problem of maximizing the minimum user signal-to-noise ratio satisfying a transmitting power constraint, an eavesdropper robust signal-to-noise ratio constraint and a unit mode constraint of an intelligent reflector element; converting the optimization problem to obtain a converted optimization problem; and solving the transformed optimization problem by adopting an alternating optimization method to obtain a final solution of the optimization problem. According to the invention, the system communication performance can be improved on the premise that the channel state information of the passive eavesdropper cannot be obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent reflective surface assisted wireless communication, and more specifically, to a robust beamforming design method for an intelligent reflective surface assisted downlink MISO system. Background Art

[0002] Common intelligent reflector-assisted confidential communication systems are based on the assumption that the base station can obtain perfect channel state information of the eavesdropper, which is unrealistic and incomplete in practice. When confidential information is intercepted by passive eavesdroppers (PEves) and active eavesdroppers (AEves) and the channel state information of the passive eavesdroppers cannot be obtained, how to improve the communication performance of the system has become a problem that needs to be solved. Summary of the invention

[0003] The present invention aims to solve the problem that the communication quality is not high in the current technology when the channel state information of the passive eavesdropper cannot be obtained, and provides a robust beamforming design method for a downlink MISO system assisted by an intelligent reflector.

[0004] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0005] The present invention proposes a robust beamforming design method for a downlink MISO system assisted by an intelligent reflector, comprising:

[0006] Establishing an intelligent reflective surface assisted downlink MISO system, the system includes 1 base station, 1 user, 1 intelligent reflective surface, K active eavesdroppers and an unknown number of passive eavesdroppers;

[0007] According to the intelligent reflector-assisted downlink MISO system, the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model are constructed;

[0008] According to the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model, an optimization problem of maximizing the minimum user signal-to-noise ratio that satisfies the transmission power constraint, the eavesdropper robust signal-to-noise ratio constraint and the unit mode constraint of the smart reflector element is established;

[0009] Transforming the optimization problem to obtain a transformed optimization problem;

[0010] The transformed optimization problem is solved by using an alternating optimization method to obtain a final solution to the optimization problem.

[0011] Preferably, constructing a user signal-to-noise ratio model according to the intelligent reflective surface-assisted downlink MISO system includes:

[0012] The user receives the signal:

[0013]

[0014] Among them, y B (t) represents the user received signal, s(t) represents the data symbol required by the user, represents the Gaussian white noise received by the user, σ B 2 is the variance, with size σ 2 ;

[0015] The user signal-to-noise ratio model is:

[0016]

[0017] Among them, SNR B represents the user signal-to-noise ratio, w is the transmit beamforming vector, M is the number of antennas, and v = [v1, v2, …, v N ] H represents the reflection beam forming vector of the smart reflector, N is the number of reflective elements of the smart reflector, represents the base station-intelligent reflector-user joint channel, G represents the base station-user channel, the smart reflector-user channel and the base station-smart reflector channel respectively, and is the channel estimation value, ΔG B , Δh d,B are channel estimation errors and are respectively confined in a bounded Frobenius norm ball.

[0018] Preferably, the eavesdropper signal-to-noise ratio model includes an active eavesdropper model:

[0019] The signal received by the active eavesdropper is:

[0020]

[0021] Among them, y ae (t) represents the active eavesdropper receiving signal, represents the Gaussian white noise received by the active eavesdropper, σ ae 2 is the variance, with size σ 2 ;

[0022] Calculate the active eavesdropper signal-to-noise ratio:

[0023]

[0024] Among them, SNR ae represents the active eavesdropper signal-to-noise ratio, represents base station-intelligent reflective surface-active eavesdropper-joint channel, are channel estimation values, ΔG ae , Δhd,ae are channel estimation errors and are respectively restricted to a bounded Frobenius norm ball; They represent the base station-active eavesdropper channel and the smart reflection surface-active eavesdropper channel respectively, and K is the number of active eavesdroppers.

[0025] Preferably, the eavesdropper signal-to-noise ratio model also includes a passive eavesdropper model:

[0026] The passive eavesdropper receives the signal:

[0027]

[0028] Among them, y pe (t) represents the passive eavesdropper receiving signal, represents the Gaussian white noise received by the passive eavesdropper, σ pe 2 is the variance, with size σ 2 ;

[0029] Calculate the passive eavesdropper signal-to-noise ratio:

[0030]

[0031] in, represents the passive eavesdropper's signal-to-noise ratio, E[·] represents the mean operation, represents base station-intelligent reflective surface-passive eavesdropper-joint channel, are channel estimation values, ΔG pe , Δh d,pe are channel estimation errors, Δh d,pe The mean is 0 and the covariance matrix is ​​H p A random vector of They represent the base station-passive eavesdropper channel and the smart reflection surface-passive eavesdropper channel respectively.

[0032] Preferably, according to the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model, establishing an optimization problem of maximizing the minimum user signal-to-noise ratio that satisfies the transmission power constraint, the eavesdropper robust signal-to-noise ratio constraint and the unit mode constraint of the smart reflective surface element comprises:

[0033]

[0034] |v n | 2 =1,n=1,2,…,N

[0035] ‖w‖ 2 ≤P

[0036] in, is the channel estimation matrix, ΔG B , ΔG ae , ΔG pe is the channel error matrix, is the channel estimation vector, v n is the nth element of the main diagonal of v, P is the transmit power threshold, Δh d,B , Δh d,ae , Δh d,pe Channel error vector; SNR ae,k represents the maximum value of the signal-to-noise ratio of k active eavesdroppers; ε d,B , ε B , ε d,ae , ε ae is the radius of the channel error norm sphere, ΔG B satisfy ||Δh d,B ||2≤ε d,B ,‖ΔG B ‖ F ≤ε B , ||Δh d,ae ||2≤ε d,ae , ‖ΔG ae ‖ F ≤ε ae , Δh d,pe The mean is 0 and the covariance is H. p Gaussian distribution, ΔG pe The mean is 0 and the covariance is G p The Gaussian distribution of N is the number of reflective elements of the smart reflective surface.

[0037] Preferably, the optimization problem is transformed to obtain the transformed optimization problem, which includes:

[0038] Introducing s, s>0, after a series of simplifications, problem P1 is equivalently transformed into problem P2:

[0039]

[0040] |v n | 2 =1,n=1,2,…,N(1d)

[0041] ‖w‖ 2 ≤P(1e)

[0042] Preferably, the transformed optimization problem is solved by an alternating optimization method to obtain a final solution to the optimization problem, which includes:

[0043] S7.1: First fix v, optimize w and s, and get w * ,s * ;

[0044] S7.2: Fix w again * ,s * , optimize v and get v * ;

[0045] S7.3: Calculate the absolute difference between the current s value and the last optimization target s value. If the absolute error is less than the preset threshold, output w * ,s * , v * As the final solution, otherwise, return to step S7.1.

[0046] Preferably, the step S7.1 comprises:

[0047] Fix v, optimize w and s, and we will optimize the constraint (1a) in problem P2 Taking the real part, we can convert it into The optimization problem P2 becomes a SOCP problem, and the optimal solution w is obtained by using the CVX toolkit. * ,s * .

[0048] Preferably, the step S7.2 comprises:

[0049] Fixed w = w * , s=s * , using SDR technology to solve:

[0050] Removing the rank-one constraint, the optimization problem P2 becomes the following form:

[0051] P3: find V

[0052] stTr(BV)≥α

[0053] Tr[(P1+P2)V]≤β

[0054] Tr(A k V)≤λ

[0055] V n,n =1,n=1,…,N+1

[0056]

[0057] in, and

[0058] The optimization problem P3 is a convex problem. We use the CVX toolkit to solve it and obtain the optimal solution v * ;

[0059] If the optimal solution v * The rank is 1, then for v * Perform eigenvalue decomposition to obtain the final optimal solution v1 of the optimization problem P1 * ; If the optimal solution v * If the rank is not 1, the random vector constructed by Gaussian randomization makes the objective function value of the optimization problem P3 converge. At this time, the Gaussian random vector is used as the final optimal solution v1 of the optimization problem P1 * .

[0060] The present invention also proposes a robust beamforming design system for a downlink MISO system assisted by an intelligent reflector, which is used to implement the above method, including:

[0061] A system establishment module, used for establishing an intelligent reflective surface assisted downlink MISO system, the system comprising 1 base station, 1 user, 1 intelligent reflective surface, K active eavesdroppers and an unknown number of passive eavesdroppers;

[0062] A model building module, used for building a user signal-to-noise ratio model and an eavesdropper signal-to-noise ratio model according to the intelligent reflector-assisted downlink MISO system;

[0063] An optimization problem establishment module is used to establish an optimization problem of maximizing the minimum user signal-to-noise ratio that satisfies the transmission power constraint, the eavesdropper robust signal-to-noise ratio constraint and the unit mode constraint of the smart reflective surface element according to the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model;

[0064] A transformation module, used to transform the optimization problem to obtain a transformed optimization problem;

[0065] The solution module is used to solve the transformed optimization problem by adopting an alternating optimization method to obtain a final solution to the optimization problem.

[0066] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0067] The present invention proposes a robust beamforming design method for a smart reflector-assisted downlink MISO system, comprising: establishing a smart reflector-assisted downlink MISO system, the system comprising 1 base station, 1 user, 1 smart reflector, K active eavesdroppers and an unknown number of passive eavesdroppers; constructing a user signal-to-noise ratio model and an eavesdropper signal-to-noise ratio model according to the smart reflector-assisted downlink MISO system; establishing an optimization problem of maximizing the minimum user signal-to-noise ratio that satisfies a transmission power constraint, an eavesdropper robust signal-to-noise ratio constraint and a unit mode constraint of a smart reflector element according to the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model; transforming the optimization problem to obtain a transformed optimization problem; solving the transformed optimization problem using an alternating optimization method to obtain a final solution to the optimization problem. The present invention can improve the communication performance of the system under the premise that the channel state information of the passive eavesdropper cannot be obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 This is a schematic flow chart of the method described in Example 1;

[0069] Figure 2 is a system block diagram of the method described in Example 2;

[0070] Figure 3 This is a schematic diagram of the structure of the system described in Example 3. DETAILED DESCRIPTION

[0071] The drawings are for illustrative purposes only and should not be construed as limiting the present patent;

[0072] In order to better illustrate the present embodiment, some parts in the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product;

[0073] It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0074] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0075] Example 1

[0076] This embodiment provides a robust beamforming design method for a downlink MISO system assisted by an intelligent reflector. Figure 1 As shown, including:

[0077] Establishing an intelligent reflective surface assisted downlink MISO system, the system includes 1 base station, 1 user, 1 intelligent reflective surface, K active eavesdroppers and an unknown number of passive eavesdroppers;

[0078] According to the intelligent reflector-assisted downlink MISO system, the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model are constructed;

[0079] According to the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model, an optimization problem of maximizing the minimum user signal-to-noise ratio that satisfies the transmission power constraint, the eavesdropper robust signal-to-noise ratio constraint and the unit mode constraint of the smart reflector element is established;

[0080] Transforming the optimization problem to obtain a transformed optimization problem;

[0081] The transformed optimization problem is solved by using an alternating optimization method to obtain a final solution to the optimization problem.

[0082] In the specific implementation process, firstly, the intelligent reflector-assisted downlink MISO system is established, and the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model are constructed. Then, based on the model, an optimization problem of maximizing the minimum user signal-to-noise ratio that satisfies the transmission power constraint, the eavesdropper robust signal-to-noise ratio constraint, and the unit mode constraint of the intelligent reflector element is established; finally, the optimization problem is transformed to obtain the transformed optimization problem, and the transformed optimization problem is solved by the alternating optimization method to obtain the final solution of the optimization problem.

[0083] Example 2

[0084] The present invention proposes a robust beamforming design method for a downlink MISO system assisted by an intelligent reflector, comprising:

[0085] Establish an intelligent reflector-assisted downlink MISO system, such as Figure 2 As shown, the system includes 1 base station, 1 user, 1 intelligent reflective surface, K active eavesdroppers and an unknown number of passive eavesdroppers;

[0086] According to the intelligent reflector-assisted downlink MISO system, the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model are constructed as follows:

[0087] The user receives the signal:

[0088]

[0089] The signal received by the active eavesdropper is:

[0090]

[0091] The passive eavesdropper receives the signal:

[0092]

[0093] Among them, s(t) represents the data symbol required by the user, Represented as Gaussian white noise at the receiver, σ B 2, σ ae 2 and σ pe 2 is the variance, and its size is σ 2 .

[0094] The user signal-to-noise ratio model is:

[0095]

[0096] Among them, SNR B represents the user signal-to-noise ratio, w is the transmit beamforming vector, M is the number of antennas, and v = [v1, v2, …, v N ] H represents the reflection beam forming vector of the smart reflector, N is the number of reflective elements of the smart reflector, represents the base station-intelligent reflector-user joint channel, G represents the base station-user channel, the smart reflector-user channel and the base station-smart reflector channel respectively, and is the channel estimation value, ΔG B , Δh d,B are channel estimation errors and are respectively confined in a bounded Frobenius norm ball.

[0097] Active eavesdropper model:

[0098]

[0099] Among them, SNR ae represents the active eavesdropper signal-to-noise ratio, represents base station-intelligent reflective surface-active eavesdropper-joint channel, are channel estimation values, ΔG ae , Δh d,ae are channel estimation errors and are respectively restricted to a bounded Frobenius norm ball; They represent the base station-active eavesdropper channel and the smart reflection surface-active eavesdropper channel respectively, and K is the number of active eavesdroppers.

[0100] Passive eavesdropper model:

[0101]

[0102] in, represents the passive eavesdropper's signal-to-noise ratio, E[·] represents the mean operation, represents base station-intelligent reflective surface-passive eavesdropper-joint channel, are channel estimation values, ΔG pe , Δhd,pe are channel estimation errors, Δh d,pe The mean is 0 and the covariance matrix is ​​H p A random vector of They represent the base station-passive eavesdropper channel and the smart reflection surface-passive eavesdropper channel respectively.

[0103] According to the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model, an optimization problem of maximizing the minimum user signal-to-noise ratio that satisfies the transmission power constraint, the eavesdropper robust signal-to-noise ratio constraint, and the unit mode constraint of the smart reflector element is established. Specifically,

[0104]

[0105] |v n | 2 =1,n=1,2,…,N

[0106] ‖w‖ 2 ≤P

[0107] in, is the channel estimation matrix, ΔG B , ΔG ae , ΔG pe is the channel error matrix, is the channel estimation vector, v n is the nth element of the main diagonal of v, P is the transmit power threshold, Δh d,B , Δh d,ae , Δh d,pe Channel error vector; SNR ae,k represents the maximum value of the signal-to-noise ratio of k active eavesdroppers; ε d,B , ε B , ε d,ae , ε ae is the radius of the channel error norm sphere, ΔG B satisfy ||Δh d,B ||2≤ε d,B ,‖ΔG B ‖ F ≤ε B , ||Δh d,ae ||2≤ε d,ae , ‖ΔG ae ‖ F ≤ε ae , Δh d,pe The mean is 0 and the covariance is H. p Gaussian distribution, ΔG pe The mean is 0 and the covariance is G p The Gaussian distribution of N is the number of reflective elements of the smart reflective surface.

[0108] The optimization problem is transformed to obtain the transformed optimization problem, which is specifically:

[0109] Introducing s, s>0, after a series of simplifications, problem P1 is equivalently transformed into problem P2:

[0110]

[0111] |v n | 2 =1,n=1,2,…,N(1d)

[0112] ‖w‖ 2 ≤P(1e)

[0113] The transformed optimization problem is solved by an alternating optimization method. When optimizing the transmit beamforming vector, the algorithm transforms it into a second-order cone programming (SOCP) problem. When optimizing the RIS phase shift vector, the semidefinite relaxation (SDR) and Gaussian randomization techniques are applied to obtain the final solution of the optimization problem, which is:

[0114] S7.1: First fix v, optimize w and s, and get w * ,s * ;

[0115] Fix v, optimize w and s, and we will optimize the constraint (1a) in problem P2 Taking the real part, we can convert it into The optimization problem P2 becomes a SOCP problem, and the optimal solution w is obtained by using the CVX toolkit. * ,s * .

[0116] S7.2: Fix w again * ,s * , optimize v and get v * ;

[0117] Fixed w = w * , s=s * , using SDR technology to solve:

[0118] Removing the rank-one constraint, the optimization problem P2 becomes the following form:

[0119] P3: find V

[0120] stTr(BV)≥α

[0121] Tr[(P1+P2)V]≤β

[0122] Tr(A k V)≤λ

[0123] V n,n =1,n=1,…,N+1

[0124]

[0125] in, and

[0127] The optimization problem P3 is a convex problem. We use the CVX toolkit to solve it and obtain the optimal solution v * ;

[0128] If the optimal solution v * The rank is 1, then for v * Perform eigenvalue decomposition to obtain the final optimal solution v1 of the optimization problem P1 * ; If the optimal solution v * If the rank is not 1, the random vector constructed by Gaussian randomization makes the objective function value of the optimization problem P3 converge. At this time, the Gaussian random vector is used as the final optimal solution v1 of the optimization problem P1 * .

[0129] S7.3: Calculate the absolute difference between the current s value and the last optimization target s value. If the absolute error is less than the preset threshold, output w * ,s * , v * As the final solution, otherwise, return to step S7.1.

[0130] This embodiment combines transmit beamforming and RIS reflection beamforming designs. Compared with the existing beamforming design under the MISO system, this embodiment takes into account the uncertainty of the base station to the communication user and AEves channels and the channel random variables from the base station to PEves obeying the Gaussian distribution, which is more in line with practical applications and has good system robustness. It also proposes the problem of maximizing the user signal-to-noise ratio and makes the maximum AEves signal-to-noise ratio and the expected value of the PEves signal-to-noise ratio less than the corresponding threshold. At the same time, under the constraints of the constant modulus constraint and the transmit beamforming vector, a single SOCP is used when processing the transmit beamforming vector, and the SDR technology and the Gaussian randomization method are used when processing the RIS unit modulus constraint, so as to quickly solve the optimization problem.

[0131] Example 3

[0132] This embodiment provides a robust beamforming design system for a downlink MISO system assisted by an intelligent reflector, which is used to implement the method described in Embodiment 1 or 2. Figure 3 As shown, including:

[0133] A system establishment module, used for establishing an intelligent reflective surface assisted downlink MISO system, the system comprising 1 base station, 1 user, 1 intelligent reflective surface, K active eavesdroppers and an unknown number of passive eavesdroppers;

[0134] A model building module, used for building a user signal-to-noise ratio model and an eavesdropper signal-to-noise ratio model according to the intelligent reflector-assisted downlink MISO system;

[0135] An optimization problem establishment module is used to establish an optimization problem of maximizing the minimum user signal-to-noise ratio that satisfies the transmission power constraint, the eavesdropper robust signal-to-noise ratio constraint and the unit mode constraint of the smart reflective surface element according to the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model;

[0136] A transformation module, used to transform the optimization problem to obtain a transformed optimization problem;

[0137] The solution module is used to solve the transformed optimization problem by adopting an alternating optimization method to obtain a final solution to the optimization problem.

[0138] The same or similar reference numerals correspond to the same or similar components;

[0139] The terms used in the drawings to describe positional relationships are only used for illustrative purposes and should not be construed as limiting this patent;

[0140] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A robust beamforming design method for a smart reflector-assisted downlink MISO system, characterized in that: include: Establishing an intelligent reflective surface assisted downlink MISO system, the system includes 1 base station, 1 user, 1 intelligent reflective surface, K active eavesdroppers and an unknown number of passive eavesdroppers; According to the intelligent reflector-assisted downlink MISO system, the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model are constructed; According to the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model, an optimization problem of maximizing the minimum user signal-to-noise ratio that satisfies the transmission power constraint, the eavesdropper robust signal-to-noise ratio constraint and the unit mode constraint of the smart reflector element is established; Transforming the optimization problem to obtain a transformed optimization problem; The transformed optimization problem is solved by using an alternating optimization method to obtain a final solution to the optimization problem.

2. The robust beamforming design method for a smart reflector-assisted downlink MISO system according to claim 1, characterized in that: The constructing of a user signal-to-noise ratio model according to the intelligent reflector-assisted downlink MISO system includes: The user receives the signal: Among them, y B (t) represents the user received signal, s(t) represents the data symbol required by the user, represents the Gaussian white noise received by the user, σ B 2 is the variance, with size σ 2 ; The user signal-to-noise ratio model is: Among them, SNR B represents the user signal-to-noise ratio, w is the transmit beamforming vector, M is the number of antennas, and v = [v1, v2, …, v N ] H represents the reflection beam forming vector of the smart reflector, N is the number of reflective elements of the smart reflector, represents the base station-intelligent reflector-user joint channel, G represents the base station-user channel, the smart reflector-user channel and the base station-smart reflector channel respectively, and is the channel estimation value, ΔG B , Δh d,B are channel estimation errors and are respectively confined in a bounded Frobenius norm ball.

3. The robust beamforming design method for a smart reflector-assisted downlink MISO system according to claim 2, characterized in that: The eavesdropper signal-to-noise ratio model includes an active eavesdropper model: The signal received by the active eavesdropper is: Among them, y ae (t) represents the active eavesdropper receiving signal, represents the Gaussian white noise received by the active eavesdropper, σ ae 2 is the variance, with size σ 2 ; Calculate the active eavesdropper signal-to-noise ratio: Among them, SNR ae represents the active eavesdropper signal-to-noise ratio, represents base station-intelligent reflective surface-active eavesdropper-joint channel, are channel estimation values, ΔG ae , Δh d,ae are channel estimation errors and are respectively restricted to a bounded Frobenius norm ball; They represent the base station-active eavesdropper channel and the smart reflection surface-active eavesdropper channel respectively, and K is the number of active eavesdroppers.

4. The robust beamforming design method for a smart reflector-assisted downlink MISO system according to claim 2, characterized in that: The eavesdropper signal-to-noise ratio model also includes a passive eavesdropper model: The passive eavesdropper receives the signal: Among them, y pe (t) represents the passive eavesdropper receiving signal, represents the Gaussian white noise received by the passive eavesdropper, σ pe 2 is the variance, with size σ 2 ; Calculate the passive eavesdropper signal-to-noise ratio: in, represents the passive eavesdropper's signal-to-noise ratio, E[·] represents the mean operation, represents base station-intelligent reflective surface-passive eavesdropper-joint channel, are channel estimation values, ΔG pe , Δh d,pe are channel estimation errors, Δh d,pe The mean is 0 and the covariance matrix is ​​H p A random vector of They represent the base station-passive eavesdropper channel and the smart reflection surface-passive eavesdropper channel respectively.

5. The robust beamforming design method for a smart reflector-assisted downlink MISO system according to claim 3 or 4, characterized in that: According to the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model, the optimization problem of maximizing the minimum user signal-to-noise ratio that satisfies the transmission power constraint, the eavesdropper robust signal-to-noise ratio constraint and the unit mode constraint of the smart reflector element is established, including: |v n | 2 =1,n=1,2,…,N ‖In‖ 2 ≤P in, is the channel estimation matrix, ΔG B , ΔG ae , ΔG pe is the channel error matrix, is the channel estimation vector, v n is the nth element of the main diagonal of v, P is the transmit power threshold, Δh d,B , Δh d,ae , Δh d,pe Channel error vector; SNR ae,k represents the maximum value of the signal-to-noise ratio of k active eavesdroppers; ε d,B , ε B , ε d,ae , ε ae is the radius of the channel error norm sphere, ΔG B satisfy ||Δh d,B ||2≤ε d,B ,‖ΔG B ‖ F ≤ε B , ||Δh d,ae ||2≤ε d,ae , ‖ΔG ae ‖ F ≤ε ae , Δh d,pe The mean is 0 and the covariance is H. p Gaussian distribution, ΔG pe The mean is 0 and the covariance is G p The Gaussian distribution of N is the number of reflective elements of the smart reflective surface.

6. The robust beamforming design method for a smart reflector-assisted downlink MISO system according to claim 5, characterized in that: The optimization problem is transformed to obtain the transformed optimization problem including: Introducing s, s>0, after a series of simplifications, problem P1 is equivalently transformed into problem P2: |v n | 2 =1,n=1,2,…,N(1d) ‖In‖ 2 ≤p(1e)。 7. The robust beamforming design method for a smart reflector-assisted downlink MISO system according to claim 6, characterized in that: The transformed optimization problem is solved by using an alternating optimization method, and the final solution of the optimization problem is obtained, including: S7.1: First fix v, optimize w and s, and get w * ,s * ; S7.2: Fix w again * ,s * , optimize v and get v * ; S7.3: Calculate the absolute difference between the current s value and the last optimization target s value. If the absolute error is less than the preset threshold, output w * ,s * , v * As the final solution, otherwise, return to step S7.

1.

8. The robust beamforming design method for a smart reflector-assisted downlink MISO system according to claim 7, characterized in that: The step S7.1 comprises: Fix v, optimize w and s, and we will optimize the constraint (1a) in problem P2 Taking the real part, we can convert it into The optimization problem P2 becomes a SOCP problem, and the optimal solution w is obtained by using the CVX toolkit. * ,s * .

9. The robust beamforming design method for a smart reflector-assisted downlink MISO system according to claim 8, characterized in that: The step S7.2 comprises: Fixed w = w * , s=s * , using SDR technology to solve: Removing the rank-one constraint, the optimization problem P2 becomes the following form: P3: find V st Tr(BW)≥α Tr[(P1+P2)V]≤β Tr(A k V)≤λ V n,n =1,n=1,…,N+1 V≥0 in, and The optimization problem P3 is a convex problem. We use the CVX toolkit to solve it and obtain the optimal solution v * ; If the optimal solution v * The rank is 1, then for v * Perform eigenvalue decomposition to obtain the final optimal solution v1 of the optimization problem P1 * ; If the optimal solution v * If the rank is not 1, the random vector constructed by Gaussian randomization makes the objective function value of the optimization problem P3 converge. At this time, the Gaussian random vector is used as the final optimal solution v1 of the optimization problem P1 * .

10. A robust beamforming design system for a downlink MISO system assisted by an intelligent reflector, used to implement the method according to any one of claims 1 to 9, characterized in that: include: A system establishment module, used for establishing an intelligent reflective surface assisted downlink MISO system, the system comprising 1 base station, 1 user, 1 intelligent reflective surface, K active eavesdroppers and an unknown number of passive eavesdroppers; A model building module, used for building a user signal-to-noise ratio model and an eavesdropper signal-to-noise ratio model according to the intelligent reflector-assisted downlink MISO system; An optimization problem establishment module is used to establish an optimization problem of maximizing the minimum user signal-to-noise ratio that satisfies the transmission power constraint, the eavesdropper robust signal-to-noise ratio constraint and the unit mode constraint of the smart reflective surface element according to the user signal-to-noise ratio model and the eavesdropper signal-to-noise ratio model; A transformation module, used to transform the optimization problem to obtain a transformed optimization problem; The solution module is used to solve the transformed optimization problem by adopting an alternating optimization method to obtain a final solution to the optimization problem.

Images