Method for maximizing minimum safety rate of RIS-assisted unmanned aerial vehicle relay system

Abstract

The invention discloses a method for maximizing the minimum safety rate of an RIS-assisted unmanned aerial vehicle relay system, which considers the fairness among legal users, and maximizes the minimum safety rate of the RIS-assisted unmanned aerial vehicle relay system under the condition of meeting the service and quality of the legal users, the maximum transmitting power of a base station, RIS phase shift and unmanned aerial vehicle position constraints. According to the method, a minimum safety rate maximization model based on the RIS auxiliary unmanned aerial vehicle relay system is established, a non-convex optimization problem is equivalently converted into a convex optimization problem by introducing auxiliary variables, continuous convex approximation and other methods, and then the minimum safety rate between legal users is maximized by using an alternating iteration algorithm. According to the method, the global fairness of resource allocation can be well realized while the service quality requirements of all legal users are ensured, and the method has better practicability and feasibility.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communication system resource allocation, and in particular, relates to a method for maximizing the minimum safety rate of a RIS-assisted unmanned aerial vehicle relay system. Background Art

[0002] With the rapid development of science and technology, wireless communication networks have gradually matured. However, due to the openness and broadcasting of wireless channels, security issues still exist. How to improve the security of information transmission in wireless communication systems has become a key issue that needs to be solved urgently. Recently, as a promising technology, Reconfigurable Intelligence Surface (RIS) has attracted widespread attention from academia and industry. The surface of RIS integrates a large number of passive reflective elements, which can intelligently reconfigure the wireless propagation environment. Each reflective element can control its amplitude and phase to reflect the incident signal. By jointly controlling the phase shift of all reflective elements of RIS, the transmission direction of the reflected signal is changed, thereby improving the security performance of the wireless communication system. In addition, drones have the characteristics of low cost, flexibility, and on-demand deployment. Deploying RIS on drones as air relays can further improve the performance of wireless communication systems. In recent years, Rate-Splitting Multiple Access (RSMA) as a new multiple access technology can not only support large-scale connections, but also achieve higher spectrum efficiency than Non-Orthogonal Multiple Access (NOMA) and Space Division Multiple Access (SDMA). By introducing RSMA into the RIS-assisted UAV relay system, the public messages sent by the transmitter are not only useful messages required by legitimate users, but can also serve as AN to confuse eavesdroppers, thereby effectively enhancing the physical layer security performance of the system.

[0003] At present, many studies have been conducted on the resource allocation problem of physical layer security in RIS-assisted UAV relay systems. ZHANG Y et al. published an article entitled "Joint Optimization for Aerial Intelligent Reflecting Surface Assisted Secure Wireless Network" in "2022IEEE 8th International Conference on Computer and Communications (ICCC), Chengdu, China, 2022: 1000-1004.", which studied the physical layer security problem in RIS-assisted UAV relay systems and maximized the average security rate of the minimum user by jointly optimizing the phase shift matrix of RIS, communication scheduling and the trajectory of the UAV. WEI W et al. published an article entitled "Secure transmission design for aerial IRS assisted wireless networks" in "IEEE Transactions on Communications, 2023, 71(6): 3528-3540.", which studied the situation where there are multiple users and multiple eavesdroppers in the RIS-assisted UAV relay system, and maximized the worst-case and security rate by jointly optimizing the hovering position of the UAV, the transmission beamforming of the base station and the phase shift of the RIS.

[0004] From the above results, it can be seen that most of the current research on physical layer security issues in RIS-assisted UAV relay systems considers the use of Time Division Multiple Access (TDMA), SDMA or NOMA protocols, and does not consider the use of RSMA protocol. In wireless security communication systems, the use of RSMA protocol can not only support multi-user access, but also effectively improve the security performance of the system. Therefore, the present invention considers the case where the base station and the legitimate user use the RSMA protocol for communication, and studies the method of maximizing the minimum security rate of the RIS-assisted UAV relay system. Summary of the invention

[0005] The present invention aims to solve the above problems of the prior art. A method for maximizing the minimum safe rate of a RIS-assisted UAV relay system is proposed. The technical solution of the present invention is as follows:

[0006] A method for maximizing the minimum safety rate of a RIS-assisted unmanned aerial vehicle relay system comprises the following steps:

[0007] Step 1: Deploy a RIS-assisted drone relay system with M reflector units. The drone carries RIS as an aerial relay to assist the base station in communicating with legitimate users. At the same time, there is an eavesdropper trying to eavesdrop on the legitimate user's data information.

[0008] Step 2: Considering the fairness and communication reliability between legitimate users, the minimum safe rate maximization problem of the RIS-assisted UAV relay system is constructed. The optimization variables are the base station transmission power P, the public rate allocation C, the RIS phase shift Θ and the UAV position w u ;

[0009] Step 3: Initialize base station transmit power P, public rate allocation C, RIS phase shift Θ and drone position w u , set the maximum number of iterations n max , convergence accuracy ε;

[0010] Step 4: Fix RIS phase shift Θ and drone position w u , solve the base station transmission power P and public rate allocation C;

[0011] Step 5: Fix the base station transmission power P, public rate allocation C and drone position w u , solve for the RIS phase shift Θ;

[0012] Step 6: Fix the base station transmit power P, public rate allocation C and RIS phase shift Θ, and solve the drone position w u ;

[0013] Step 7: Alternately iterate and optimize the base station transmit power P, public rate allocation C, RIS phase shift Θ and drone position w u , until the minimum safe rate error value is less than the threshold or the maximum number of iterations is reached.

[0014] Furthermore, in step 1, the drone relay system model is as follows:

[0015] Consider a RIS-assisted drone relay secure communication system. The model consists of a base station, a drone equipped with RIS, K legitimate users and an eavesdropper. The base station, legitimate users and eavesdroppers are all equipped with single antennas. The base station and legitimate users communicate using the RSMA protocol.

[0016] The transmitted signal at the base station is expressed as:

[0017]

[0018] s c and k denote the public message transmitted by the base station and the private message of the kth legal user, respectively, c and pk They are c and k Therefore, the signals received at the legitimate user k and the eavesdropper are expressed as:

[0019]

[0020] Among them, h k 、h e and h r They represent the channel gains from RIS to legitimate user k, RIS to eavesdropper, and base station to RIS, respectively. denote the additive Gaussian white noise at the legitimate user k and the eavesdropper respectively;

[0021] According to the RSMA principle, the instantaneous signal-to-interference-noise ratio of decoding public messages and private messages at the legitimate user k is expressed as:

[0022]

[0023] in, is the noise variance at the legitimate user k, so the instantaneous achievable rate of decoding public messages and private messages by the legitimate user k under unit bandwidth is expressed as:

[0024] R kc =log2(1+Γ c,k )

[0025] R kp =log2(1+Γ p,k )

[0026] To ensure that the public message can be successfully decoded by all users, the actual transmission rate of the public message R c The minimum achievable rate of public messages among all legitimate users is not exceeded, that is, R c Satisfy R c ≤min k∈K R kc In addition, R c Shared by all legitimate users, each legitimate user k is assigned to R c Part of C k ; Common rate allocation vector C1,C2,...,C K Should meet ∑ k∈K C k =R c ; Therefore, the total achievable rate of legitimate user k is expressed as:

[0027] R k,tot =C k +R kp

[0028] The achievable eavesdropping rate of an eavesdropper decoding public messages and private messages of legitimate user k is expressed as:

[0029] R ec =log2(1+Γ ec )

[0030]

[0031] in, and They represent the signal-to-interference-to-noise ratio of the public message decoded at the eavesdropper and the private message of the legitimate user k, respectively, and are expressed as:

[0032]

[0033] in, is the noise variance at the eavesdropper; assuming that the public message achievable eavesdropping rate at the eavesdropper is evenly distributed to each legitimate user, then the total achievable eavesdropping rate of eavesdropping legitimate user k can be expressed as:

[0034]

[0035] Finally, the safe rate of legitimate user k is defined as follows:

[0036]

[0037] Where [x] + =max{x,0},[·] + It can be omitted because at the optimal solution, the security rate R corresponding to each legitimate user s,k At least 0.

[0038] Furthermore, in step 2, the minimum safety rate maximization optimization problem constructed is:

[0039]

[0040] C7:0≤x u ≤X,0≤y u ≤Y

[0041] Among them, constraint C1 is the minimum rate requirement constraint for each legal user, represents the minimum rate requirement of each legal user, constraint C2 is the maximum transmit power constraint at the base station, and p maxis the maximum transmit power at the base station, constraint C3 ensures that each legitimate user can decode the public stream, constraint C4 ensures that the public stream cannot be completely decoded by the eavesdropper, constraint C5 indicates that the public rate assigned to each legitimate user is non-negative; constraint C6 represents the phase shift constraint of each reflection unit of RIS, θ m represents the reflection phase of each reflection unit of RIS; constraint C7 indicates that the UAV is deployed within a certain area, x u Indicates the horizontal coordinate of the drone's position, y u Indicates the horizontal and vertical coordinates of the drone's position, X indicates the maximum value of the horizontal abscissa of the drone's position, and Y indicates the maximum value of the horizontal and vertical coordinates of the drone's position.

[0042] Furthermore, in step 4, the steps of obtaining the joint optimization sub-problem of base station transmission power P and public rate allocation C and converting it into an equivalent convex optimization problem are as follows:

[0043] First, fix the RIS phase shift Θ and the drone position w u , we get the joint optimization sub-problem of base station transmission power P and public rate allocation C:

[0044]

[0045] stC1~C5

[0046]

[0047] Where τ1 is the introduced slack variable, given the RIS phase shift Θ and the UAV position w u In the case of , C2 and C5 are affine with respect to the optimization variables P and C, while C1, C3, C4 and C8 are all non-convex. Therefore, this subproblem is a non-convex optimization problem.

[0048] Let R s,k =C k +(f kp -g kp )-(f ec -g ec ) / K-(f ep -g ep ), R kc =f kc -g kc ; Among them, f kc , g kc 、f kp , g kp 、f ec , g ec 、f ep , g ep They are:

[0049]

[0050] f kc , g kc They are R kc The first and second terms after transformation into the subtraction form of two functions, f kp , g kp They are R kp The first and second terms after transformation into the subtraction form of two functions, f ec , g ec They are R ec The first and second terms after transformation into the subtraction form of two functions, f ep , g ep They are R ep The first and second terms after transformation into the subtraction form of two functions.

[0051] Then, the optimization problem P2 can be rewritten as:

[0052]

[0053] stC2、C5

[0054]

[0055] Using the SCA method, g kp , g kc 、f ec and f ep At the initial feasible point P (n) Perform a first-order Taylor expansion at The optimization problem P 2.1 g kp , g kc 、f ec and f ep By replacing their upper bounds respectively, we can get an equivalent convex optimization problem:

[0056]

[0057] stC2、C5

[0058]

[0059] Furthermore, in step 5, the method of obtaining the RIS phase shift θ optimization subproblem and converting it into an equivalent convex optimization problem is as follows:

[0060] First, fix the base station transmission power P, public rate allocation C and drone position w u , let the channel gain η k =|v H Hk | 2 , η e =|v H H e | 2 ,in For the introduced auxiliary variables, the RIS phase shift Θ optimization subproblem is obtained:

[0061]

[0062] st C1, C3, C4, C6

[0063] The objective function and constraints C1, C3, and C4 of the optimization problem P3 are all non-convex with respect to the variable v and need to be processed. kc , R kp and R ec , R ep Lower bound and upper bound Replacement;

[0064] Therefore, the objective function can be rewritten as:

[0065]

[0066] The optimization problem P3 can be equivalently converted into a convex optimization subproblem:

[0067] P 3.1 :maxτ2

[0068] v,τ2

[0069]

[0070]

[0071] Among them, τ2 is the introduced slack variable.

[0072] Furthermore, in step 6, the drone position w is obtained u The steps to optimize the subproblem and convert it into an equivalent convex optimization problem are:

[0073] First, fix the base station transmit power P, public rate allocation C and RIS phase shift Θ to obtain the drone position w u Optimization sub-problem:

[0074]

[0075] stC1, C3, C4, C7

[0076] Since the objective function and constraints C1, C3, and C4 in the optimization problem P4 are related to the optimization variable w uThey are all non-convex, so the optimization problem P4 is a non-convex optimization problem and cannot be solved directly, but needs to be processed;

[0077] For ease of processing, R kc , R kp and R ec , R ep Simplifying, we have:

[0078]

[0079] Among them, d r d k d e They represent the distance between the base station and RIS, the distance between RIS and the legitimate user k, and the distance between RIS and the eavesdropper respectively; β0 represents the path loss index; B kc , D kc R kc The second term of the numerator and denominator in the simplified fraction; B kp , D kp R kp The second term of the numerator and denominator in the simplified fraction; B ec , D ec R ec The second term of the numerator and denominator in the simplified fraction; B ep , D ep R ep The second term in the numerator and denominator of the simplified fraction. B kc , B kp , B ec , B ep and D kc , D kp , D ec , D ep Respectively expressed as:

[0080]

[0081]

[0082] because and The non-convexity of The optimization problem P4 can be transformed into:

[0083]

[0084] s7C

[0085]

[0086] Among them, the optimization problem P 4.1 The objective function can be written as:

[0087]

[0088] Since in the iterative process, the optimization variable w u Changes in φ will also affect r ,φ k and φ e Therefore, B kc , B kp , B ec , B ep and D kc , D kp , D ec , D ep It is about optimizing the variable w u complex function; in order to make the problem easier to handle, kc , B kp , B ec , B ep and D kc , D kp , D ec , D ep Use their upper bounds and To replace:

[0089] B kc , B kp , B ec , B ep and D kc , D kp , D ec , D ep Substituting the upper bound of into the constraint C3″′ and the objective function, we can get:

[0090]

[0091] because is a convex function minus two convex functions, so it is still non-convex; use the SCA method on it by A first-order Taylor expansion will be Use its lower bound replace:

[0092]

[0093] in, Express At the initial feasible point The lower bound is obtained by performing a first-order Taylor expansion at .

[0094] therefore can be rewritten as:

[0095]

[0096] For the inequality The right side of the inequality is a convex function with respect to the variable U, so the inequality is a non-convex constraint. Using the SCA method, at the initial feasible point Perform a first-order Taylor expansion at , and we get:

[0097]

[0098] Therefore, the inequality can be rewritten as a convex constraint:

[0099]

[0100] Optimization Problem P 4.1 The constraints C10′, C10″ and C11′, C11″ are also non-convex and need to be processed. The non-convex terms in C10′ and C10″ are obtained by first-order Taylor expansion. The upper bound of and the non-convex terms in C11′ and C11″ The lower bound

[0101] In addition, due to the constraints C11′, C11″ For a concave function, the values ​​of and Perform a first-order Taylor expansion at:

[0102]

[0103] Therefore, the optimization problem P 4.1 It can be equivalently rewritten as a convex optimization subproblem:

[0104]

[0105] C7,C17

[0106] Among them, τ3 is the introduced slack variable.

[0107] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a method for maximizing the minimum safety rate of a RIS-assisted unmanned aerial vehicle relay system as described in any one of the items is implemented.

[0108] A non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the minimum safety rate maximization method of a RIS-assisted unmanned aerial vehicle relay system as described in any one of the items.

[0109] A computer program product comprises a computer program, wherein when the computer program is executed by a processor, the method for maximizing the minimum safety rate of a RIS-assisted unmanned aerial vehicle relay system as described in any one of the items is implemented.

[0110] The advantages and beneficial effects of the present invention are as follows:

[0111] The present invention combines RIS, UAV and RSMA, and proposes a problem of jointly optimizing base station transmission power, public rate allocation, RIS phase shift and UAV position. Since the system minimum safe rate maximization optimization problem is a non-convex problem, how to give the optimal solution is a difficult problem. Under the constraints of base station transmission power, RIS phase shift and UAV position, the present invention introduces slack variables, first-order Taylor expansion and other methods to transform the objective function and constraints into a convex optimization problem, and then solves the maximum minimum safe rate of the system. The innovation of the present invention is to apply the RSMA technology with multiple performance advantages in step 1 to the RIS-assisted UAV relay system, and on the basis of considering the fairness between legitimate users, study its physical layer security issues, and finally propose a method for maximizing the minimum safe rate. The cleverness of the present invention is that the RSMA protocol is used to enhance the communication security. In the RIS-assisted UAV relay system, the RSMA protocol is used, and the public message is not only a useful message required by the legitimate user, but also can be used as interference to confuse the eavesdropper. The flexible management of public messages by RSMA greatly reduces the complexity while enhancing the system security. BRIEF DESCRIPTION OF THE DRAWINGS

[0112] Figure 1 It is a schematic diagram of a system model of a RIS-assisted UAV relay system established in a preferred embodiment of the present invention;

[0113] Figure 2 The overall flow chart of the method for maximizing the minimum safe rate is proposed for the present invention;

[0114] Figure 3 This is a specific implementation flow chart of the method of the present invention using an alternating iterative optimization algorithm. DETAILED DESCRIPTION

[0115] The following will describe the technical solutions in the embodiments of the present invention in detail in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention.

[0116] The technical solution of the present invention to solve the above technical problems is:

[0117] Figure 3 A method for maximizing the minimum safety rate of a RIS-assisted unmanned aerial vehicle relay system is disclosed. The method comprises the following steps:

[0118] Step 1: Deploy a RIS-assisted drone relay system with M reflection units. The drone carries RIS as an aerial relay to assist the base station in communicating with legitimate users. At the same time, there is an eavesdropper trying to eavesdrop on the legitimate user's data information.

[0119] Step 2: Considering fairness and communication reliability between legitimate users, the minimum safe rate maximization problem of the RIS-assisted UAV relay system is constructed. The optimization variables are the base station transmission power P, the public rate allocation C, the RIS phase shift Θ and the UAV position w u ;

[0120] Step 3: Initialize base station transmit power P, public rate allocation C, RIS phase shift Θ and drone position w u , set the maximum number of iterations n max , convergence accuracy ε;

[0121] Step 4: Fix RIS phase shift Θ and drone position w u , solve the base station transmission power P and public rate allocation C;

[0122] Step 5: Fix the base station transmission power P, public rate allocation C and drone position w u , solve for the RIS phase shift Θ;

[0123] Step 6: Fix the base station transmit power P, public rate allocation C and RIS phase shift Θ, and solve the drone position w u ;

[0124] Step 7: Alternately iterate and optimize the base station transmit power P, public rate allocation C, RIS phase shift Θ and drone position w u , until the minimum safe rate error value is less than the threshold or the maximum number of iterations is reached.

[0125] Furthermore, in the first step, the system model is as follows:

[0126] Consider a RIS-assisted drone relay secure communication system. The system model consists of a base station, a drone equipped with RIS, K legitimate users and an eavesdropper. The base station, legitimate users and eavesdroppers are all equipped with a single antenna, and the base station and legitimate users communicate using the RSMA protocol.

[0127] The transmitted signal at the base station can be expressed as:

[0128]

[0129] s c and k denote the public message transmitted by the base station and the private message of the kth legal user, respectively, c and p k They are c and k Therefore, the signals received at the legitimate user k and the eavesdropper can be expressed as:

[0130]

[0131] Among them, h k 、h e and h r They represent the channel gains from RIS to legitimate user k, RIS to eavesdropper, and base station to RIS, respectively. represent the additive Gaussian white noise at the legitimate user k and the eavesdropper respectively.

[0132] According to the RSMA principle, the instantaneous signal-to-interference-noise ratio of decoding public messages and private messages at the legitimate user k can be expressed as:

[0133]

[0134] in, is the noise variance at the legitimate user k. Therefore, the instantaneous achievable rate of decoding public messages and private messages by the legitimate user k under unit bandwidth can be expressed as:

[0135] R kc =log2(1+Γ c,k )

[0136] R kp =log2(1+Γ p,k )

[0137] It should be noted that in order to ensure that the public message can be successfully decoded by all users, the actual transmission rate of the public message R c The minimum achievable rate of public messages at all legitimate users should not exceed R c Satisfy R c ≤min k∈K R kc In addition, R c should be shared by all legitimate users, and each legitimate user k is assigned to R c Part of C k The common rate allocation vectors C1, C2, ..., C K Should meet ∑ k∈K C k=R c Therefore, the total achievable rate of legitimate user k can be expressed as:

[0138] R k,tot =C k +R kp

[0139] The achievable eavesdropping rate of an eavesdropper decoding public messages and private messages of legitimate user k can be expressed as:

[0140] R ec =log2(1+Γ ec )

[0141]

[0142] in, and They represent the signal-to-interference-to-noise ratio of the public message decoded at the eavesdropper and the private message of the legitimate user k, respectively, and can be expressed as:

[0143]

[0144] in, is the noise variance at the eavesdropper. Without loss of generality, assuming that the public message achievable eavesdropping rate at the eavesdropper is evenly distributed to each legitimate user, then the total achievable eavesdropping rate of eavesdropping legitimate user k can be expressed as:

[0145]

[0146] Finally, the safe rate of legitimate user k is defined as follows:

[0147]

[0148] Where [x] + =max{x,0}. [·] + It can be omitted because at the optimal solution, the security rate R corresponding to each legitimate user s,k At least 0.

[0149] Furthermore, in the second step, the minimum safe rate maximization optimization problem constructed is:

[0150]

[0151] C7:0≤x u ≤X,0≤y u ≤Y

[0152] Among them, constraint C1 is the minimum rate requirement constraint for each legal user, represents the minimum rate requirement of each legal user. Constraint C2 is the maximum transmit power constraint at the base station, p max is the maximum transmit power at the base station. Constraint C3 ensures that each legitimate user can decode the public stream. Constraint C4 ensures that the public stream cannot be fully decoded by an eavesdropper. Constraint C5 indicates that the public rate assigned to each legitimate user is non-negative; Constraint C6 represents the phase shift constraint of each reflection unit of RIS, θ m represents the reflection phase of each reflection unit of RIS; constraint C7 indicates that the UAV is deployed within a certain area, x u Indicates the horizontal coordinate of the drone's position, y u Indicates the horizontal and vertical coordinates of the drone's position, X indicates the maximum value of the horizontal abscissa of the drone's position, and Y indicates the maximum value of the horizontal and vertical coordinates of the drone's position.

[0153] Furthermore, in the third step, the base station transmit power P, the public rate allocation C, the RIS phase shift θ and the drone position w are initialized. u , set the maximum number of iterations n max , convergence accuracy ε;

[0154] Furthermore, in the fourth step, the method of obtaining the joint optimization subproblem of base station transmission power P and public rate allocation C and converting it into an equivalent convex optimization problem is as follows:

[0155] First, fix the RIS phase shift Θ and the drone position w u , we get the joint optimization sub-problem of base station transmission power P and public rate allocation C:

[0156]

[0157] stC1~C5

[0158]

[0159] where τ1 is the introduced slack variable. Note that for a given RIS phase shift Θ and drone position w u In this case, C2 and C5 are affine with respect to the optimization variables P and C, while C1, C3, C4 and C8 are all non-convex. Therefore, this subproblem is a non-convex optimization problem.

[0160] Let R s,k =C k +(f kp -g kp )-(f ec -g ec ) / K-(f ep -g ep ), R kc =f kc -gkc Among them, f kc , g kc 、f kp , g kp 、f ec , g ec 、f ep , g ep They are:

[0161]

[0162] f kc , g kc They are R kc The first and second terms after transformation into the subtraction form of two functions, f kp , g kp They are R kp The first and second terms after transformation into the subtraction form of two functions, f ec , g ec They are R ec The first and second terms after transformation into the subtraction form of two functions, f ep , g ep They are R ep The first and second terms after transformation into the subtraction form of two functions.

[0163] Then, the optimization problem P2 can be rewritten as:

[0164]

[0165] stC2、C5

[0166]

[0167] Using the SCA method, g kp , g kc 、f ec and f ep At the initial feasible point P (n) Perform a first-order Taylor expansion at and The optimization problem P 2.1 g kp , g kc 、f ec and f ep By replacing their upper bounds respectively, we can get an equivalent convex optimization problem:

[0168]

[0169] stC2、C5

[0170]

[0171] Furthermore, in the fifth step, the RIS phase shift θ optimization subproblem is obtained and converted into an equivalent convex optimization problem as follows:

[0172] First, fix the base station transmission power P, public rate allocation C and drone position w u , let the channel gain η k =|v H H k | 2 , η e =|v H H e | 2 ,in is the auxiliary variable introduced. The RIS phase shift Θ optimization subproblem is obtained:

[0173]

[0174] stC1, C3, C4, C6

[0175] Note that the objective function of the optimization problem P3 and the constraints C1, C3, and C4 are all non-convex with respect to the variable v, which needs to be processed. kc , R kp and R ec , R ep Lower bound and upper bound Alternative.

[0176] Therefore, the objective function can be rewritten as:

[0177]

[0178] The optimization problem P3 can be equivalently converted into a convex optimization subproblem:

[0179]

[0180] sC6

[0181]

[0182] Among them, τ2 is the introduced slack variable.

[0183] Furthermore, in the sixth step, the drone position w is obtained u The method to optimize the subproblem and transform it into an equivalent convex optimization problem is:

[0184] First, fix the base station transmit power P, public rate allocation C and RIS phase shift Θ to obtain the drone position w u Optimization sub-problem:

[0185]

[0186] stC1, C3, C4, C7

[0187] Since the objective function and constraints C1, C3, and C4 in the optimization problem P4 are related to the optimization variable w u They are all non-convex, so the optimization problem P4 is a non-convex optimization problem and cannot be solved directly, but needs to be processed.

[0188] For ease of processing, R kc , R kp and R ec , R ep Simplifying, we have:

[0189]

[0190]

[0191] Among them, d r d k d e They represent the distance between the base station and RIS, the distance between RIS and the legitimate user k, and the distance between RIS and the eavesdropper respectively; β0 represents the path loss index; B kc , D kc R kc The second term of the numerator and denominator in the simplified fraction; B kp , D kp R kp The second term of the numerator and denominator in the simplified fraction; B ec , D ec R ec The second term of the numerator and denominator in the simplified fraction; B ep , D ep R ep The second term in the numerator and denominator of the simplified fraction. B kc , B kp , B ec , B ep and D kc , D kp , D ec , D ep Respectively expressed as:

[0192]

[0193] because and The non-convexity of The optimization problem P4 can be transformed into:

[0194]

[0195] s7C

[0196]

[0197] Among them, the optimization problem P 4.1 The objective function can be written as:

[0198]

[0199] Since in the iterative process, the optimization variable w u Changes in φ will also affect r ,φ k and φ e Therefore, B kc , B kp , B ec , B ep and D kc , D kp , D ec , D ep It is about optimizing the variable w u In order to make the problem easier to handle, kc , B kp , B ec , B ep and D kc , D kp , D ec , D ep Use their upper bounds and To replace:

[0200] B kc , B kp , B ec , B ep and D kc , D kp , D ec , D ep Substituting the upper bound of into the constraint C3″′ and the objective function, we can get:

[0201]

[0202] because is a convex function minus two convex functions, so it is still non-convex. Apply the SCA method to it by starting at the initial feasible point A first-order Taylor expansion will be Use its lower bound replace:

[0203]

[0204] in, Express At the initial feasible point The lower bound is obtained by performing a first-order Taylor expansion at .

[0205] therefore can be rewritten as:

[0206]

[0207] For the inequality The right side of the inequality is a convex function with respect to the variable U, so the inequality is a non-convex constraint. Using the SCA method, at the initial feasible point Perform a first-order Taylor expansion at , and we get:

[0208]

[0209] Therefore, the inequality can be rewritten as a convex constraint:

[0210]

[0211] Optimization Problem P 4.1 The constraints C10′, C10″ and C11′, C11″ are also non-convex and need to be processed. Use the first-order Taylor expansion to get the non-convex terms in C10′ and C10″ The upper bound of and the non-convex terms in C11′ and C11″ The lower bound

[0212] In addition, due to the constraints C11′, C11″ and For a concave function, the values ​​of and Perform a first-order Taylor expansion at:

[0213]

[0214] Therefore, the optimization problem P 4.1 It can be equivalently rewritten as a convex optimization subproblem:

[0215]

[0216] C7,C17

[0217] Among them, τ3 is the introduced slack variable.

[0218] Furthermore, in the seventh step, the base station transmission power P, the public rate allocation C, the RIS phase shift θ and the drone position w are alternately iterated and optimized. u , until the minimum safe rate error value is less than the threshold or the maximum number of iterations is reached.

[0219] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0220] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0221] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0222] The above embodiments should be understood to be only used to illustrate the present invention and not to limit the protection scope of the present invention. After reading the contents of the present invention, technicians can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. A method for maximizing the minimum safe rate of a RIS-assisted UAV relay system, characterized in that: The following steps are involved: Step 1: Deploy a RIS-assisted drone relay system with M reflector units. The drone carries RIS as an aerial relay to assist the base station in communicating with legitimate users. At the same time, there is an eavesdropper trying to eavesdrop on the legitimate user's data information. Step 2: Considering the fairness and communication reliability between legitimate users, the minimum safe rate maximization problem of the RIS-assisted UAV relay system is constructed. The optimization variables are the base station transmission power P, the public rate allocation C, the RIS phase shift Θ and the UAV position w u ; Step 3: Initialize base station transmit power P, public rate allocation C, RIS phase shift Θ and drone position w u , set the maximum number of iterations n max , convergence accuracy ε; Step 4: Fix RIS phase shift Θ and drone position w u , solve the base station transmission power P and public rate allocation C; Step 5: Fix the base station transmission power P, public rate allocation C and drone position w u , solve for the RIS phase shift Θ; Step 6: Fix the base station transmit power P, public rate allocation C and RIS phase shift Θ, and solve the drone position w u ; Step 7: Alternately iterate and optimize the base station transmit power P, public rate allocation C, RIS phase shift Θ and drone position w u , until the minimum safe rate error value is less than the threshold or the maximum number of iterations is reached.

2. The RIS-assisted UAV relay system minimum safety rate maximization method according to claim 1 is characterized in that: In step 1, the drone relay system model is as follows: Consider a RIS-assisted drone relay secure communication system. The model consists of a base station, a drone equipped with RIS, K legitimate users and an eavesdropper. The base station, legitimate users and eavesdroppers are all equipped with single antennas. The base station and legitimate users communicate using the RSMA protocol. The transmitted signal at the base station is expressed as: s c and k denote the public message transmitted by the base station and the private message of the kth legal user, respectively, c and p k They are c and k Therefore, the signals received at the legitimate user k and the eavesdropper are expressed as: Among them, h k 、h e and h r They represent the channel gains from RIS to legitimate user k, RIS to eavesdropper, and base station to RIS, respectively. denote the additive Gaussian white noise at the legitimate user k and the eavesdropper respectively; According to the RSMA principle, the instantaneous signal-to-interference-noise ratio of decoding public messages and private messages at the legitimate user k is expressed as: in, is the noise variance at the legitimate user k, so the instantaneous achievable rate of decoding public messages and private messages by the legitimate user k under unit bandwidth is expressed as: R kc =log2(1+Γ c,k ) R kp =log2(1+Γ p,k ) To ensure that the public message can be successfully decoded by all users, the actual transmission rate of the public message R c The minimum achievable rate of public messages among all legitimate users is not exceeded, that is, R c Satisfy R c ≤min k∈K R kc In addition, R c Shared by all legitimate users, each legitimate user k is assigned to R c Part of C k ; Common rate allocation vector C1,C2,...,C K Should satisfy k∈K C k =R c ; Therefore, the total achievable rate of legitimate user k is expressed as: R k,tot =C k +R kp The achievable eavesdropping rate of an eavesdropper decoding public messages and private messages of legitimate user k is expressed as: R ec =log2(1+Γ ec ) in, and They represent the signal-to-interference-to-noise ratio of the public message decoded at the eavesdropper and the private message of the legitimate user k, respectively, and are expressed as: in, is the noise variance at the eavesdropper; assuming that the public message achievable eavesdropping rate at the eavesdropper is evenly distributed to each legitimate user, then the total achievable eavesdropping rate of eavesdropping legitimate user k can be expressed as: Finally, the safe rate of legitimate user k is defined as follows: Where [x] + =max{x,0},[·] + It can be omitted because at the optimal solution, the security rate R corresponding to each legitimate user s,k At least 0.

3. The RIS-assisted UAV relay system minimum safety rate maximization method according to claim 2 is characterized in that: In step 2, the minimum safety rate maximization optimization problem constructed is: C7:0≤x u ≤X,0≤y u ≤Y Among them, constraint C1 is the minimum rate requirement constraint for each legal user, represents the minimum rate requirement of each legal user, constraint C2 is the maximum transmit power constraint at the base station, and p max is the maximum transmit power at the base station, constraint C3 ensures that each legitimate user can decode the public stream, constraint C4 ensures that the public stream cannot be completely decoded by the eavesdropper, constraint C5 indicates that the public rate assigned to each legitimate user is non-negative; constraint C6 represents the phase shift constraint of each reflection unit of RIS, θ m represents the reflection phase of each reflection unit of RIS; constraint C7 indicates that the UAV is deployed within a certain area, x u Indicates the horizontal coordinate of the drone's position, y u Indicates the horizontal and vertical coordinates of the drone's position, X indicates the maximum value of the horizontal abscissa of the drone's position, and Y indicates the maximum value of the horizontal and vertical coordinates of the drone's position.

4. The RIS-assisted UAV relay system minimum safety rate maximization method according to claim 3 is characterized in that: In step 4, the steps of obtaining the joint optimization sub-problem of base station transmission power P and public rate allocation C and converting it into an equivalent convex optimization problem are as follows: First, fix the RIS phase shift Θ and the drone position w u , we get the joint optimization sub-problem of base station transmission power P and public rate allocation C: stC1~C5 Where τ1 is the introduced slack variable, given the RIS phase shift Θ and the UAV position w u In the case of , C2 and C5 are affine with respect to the optimization variables P and C, while C1, C3, C4 and C8 are all non-convex. Therefore, this subproblem is a non-convex optimization problem. Let R s,k =C k +(f kp -g kp )-(f ec -g ec ) / K-(f ep -g ep ), R kc =f kc -g kc ; Among them, f kc , g kc 、f kp , g kp 、f ec , g ec 、f ep , g ep They are: f kc , g kc They are R kc The first and second terms after transformation into the subtraction form of two functions, f kp , g kp They are R kp The first and second terms after transformation into the subtraction form of two functions, f ec , g ec They are R ec The first and second terms after transformation into the subtraction form of two functions, f ep , g ep They are R ep The first and second terms after transformation into the subtraction form of two functions. Then, the optimization problem P2 can be rewritten as: stC2、C5 Using the SCA method, g kp , g kc 、f ec and f ep At the initial feasible point P (n) Perform a first-order Taylor expansion at and The optimization problem P 2.1 g kp , g kc 、f ec and f ep By replacing their upper bounds respectively, we can get an equivalent convex optimization problem: stC2、C5 5. The RIS-assisted UAV relay system minimum safety rate maximization method according to claim 4 is characterized in that: In step 5, the method of obtaining the RIS phase shift θ optimization subproblem and converting it into an equivalent convex optimization problem is as follows: First, fix the base station transmission power P, public rate allocation C and drone position w u , let the channel gain η k =|v H H k | 2 , η e =|v H H e | 2 ,in For the introduced auxiliary variables, the RIS phase shift Θ optimization subproblem is obtained: stC1, C3, C4, C6 The objective function and constraints C1, C3, and C4 of the optimization problem P3 are all non-convex with respect to the variable v and need to be processed. kc , R kp and R ec , R ep Lower bound and upper bound Replacement; Therefore, the objective function can be rewritten as: The optimization problem P3 can be equivalently converted into a convex optimization subproblem: sC6 Among them, τ2 is the introduced slack variable.

6. The RIS-assisted UAV relay system minimum safety rate maximization method according to claim 5, characterized in that: In step 6, the drone position w is obtained. u The steps to optimize the subproblem and convert it into an equivalent convex optimization problem are: First, fix the base station transmit power P, public rate allocation C and RIS phase shift Θ to obtain the drone position w u Optimization sub-problem: stC1, C3, C4, C7 Since the objective function and constraints C1, C3, and C4 in the optimization problem P4 are related to the optimization variable w u They are all non-convex, so the optimization problem P4 is a non-convex optimization problem and cannot be solved directly, but needs to be processed; For ease of processing, R kc , R kp and R ec , R ep Simplifying, we have: Among them, d r d k d e They represent the distance between the base station and RIS, the distance between RIS and the legitimate user k, and the distance between RIS and the eavesdropper respectively; β0 represents the path loss index; B kc , D kc R kc The second term of the numerator and denominator in the simplified fraction; B kp , D kp R kp The second term of the numerator and denominator in the simplified fraction; B ec , D ec R ec The second term of the numerator and denominator in the simplified fraction; B ep , D ep R ep The second term in the numerator and denominator of the simplified fraction. B kc , B kp , B ec , B ep and D kc , D kp , D ec , D ep Respectively expressed as: because and The non-convexity of The optimization problem P4 can be transformed into: s7C Among them, the optimization problem P 4.1 The objective function can be written as: Since in the iterative process, the optimization variable w u Changes in φ will also affect r ,φ k and φ e Therefore, B kc , B kp , B ec , B ep and D kc , D kp , D ec , D ep It is about optimizing the variable w u complex function; in order to make the problem easier to handle, kc , B kp , B ec , B ep and D kc , D kp , D ec , D ep Use their upper bounds and To replace: B kc , B kp , B ec , B ep and D kc , D kp , D ec , D ep Substituting the upper bound of into the constraint C3″′ and the objective function, we can get: because is a convex function minus two convex functions, so it is still non-convex; use the SCA method on it by A first-order Taylor expansion will be Use its lower bound replace: in, Express At the initial feasible point The lower bound is obtained by performing a first-order Taylor expansion at . therefore can be rewritten as: For the inequality The right side of the inequality is a convex function with respect to the variable U, so the inequality is a non-convex constraint. Using the SCA method, at the initial feasible point Perform a first-order Taylor expansion at, and we get: Therefore, the inequality can be rewritten as a convex constraint: Optimization Problem P 4.1 The constraints C10′, C10″ and C11′, C11″ are also non-convex and need to be processed. The non-convex terms in C10′ and C10″ are obtained by first-order Taylor expansion. The upper bound of and the non-convex terms in C11′ and C11″ The lower bound In addition, due to the constraints C11′, C11″ and For a concave function, the values ​​of and Perform a first-order Taylor expansion at: Therefore, the optimization problem P 4.1 It can be equivalently rewritten as a convex optimization subproblem: C7,C17 Among them, τ3 is the introduced slack variable.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the minimum safety rate maximization method of the RIS-assisted UAV relay system is implemented as described in any one of claims 1 to 6.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for maximizing the minimum safety rate of a RIS-assisted UAV relay system as described in any one of claims 1 to 6 is implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for maximizing the minimum safety rate of a RIS-assisted UAV relay system as described in any one of claims 1 to 6 is implemented.

Images