A method for safe communication of unmanned aerial vehicles under the assistance of intelligent reflecting surfaces

Abstract

The application relates to a kind of unmanned plane safety communication methods under the assistance of intelligent reflecting surface, the method is: constructing the downlink communication system consisting of an unmanned plane, an intelligent reflecting surface, M legitimate users and an active eavesdropper;Set target problem P1 as maximizing system average secrecy rate by jointly optimizing user scheduling, intelligent reflecting surface reflection phase shift and unmanned plane three-dimensional trajectory;Given the reflection phase shift of intelligent reflecting surface and the three-dimensional trajectory of unmanned plane input into the optimization of target problem P1 to obtain the optimal unmanned plane user scheduling;The optimal unmanned plane user scheduling and the three-dimensional trajectory given are input into the optimization of target problem P1 to obtain the optimal reflection phase shift of intelligent reflecting surface;The optimal user scheduling and the optimal reflection phase shift of intelligent reflecting surface are input into the optimization of target problem P1 to obtain the optimal three-dimensional trajectory;The method can maximize the average secrecy rate of system while ensuring the quality of service of legitimate user, and avoid user information from being eavesdropped.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) communication technology, and in particular to a secure communication method for UAVs assisted by a smart reflective surface. Background Technology

[0002] Faced with the ever-increasing number of users and the growing capacity of wireless networks, drones can effectively solve this problem, increasing the information transmission capacity of communication systems. Unmanned Aerial Vehicles (UAVs), as aerial communication platforms, possess high mobility, providing effective and reliable coverage and improving the energy efficiency of wireless communication. They can adapt to various communication environments by adjusting their position, and therefore can be deployed as mobile base stations, mobile relays, and aerial users, demonstrating advantages that traditional static base stations lack. In particular, as aerial base stations, drones can shorten the communication distance with ground users, establish strong line-of-sight (LOS) links, and provide efficient data services. Therefore, drones offer broad potential applications in 5G and beyond communication technologies.

[0003] A key challenge facing drone communication networks is that air-to-ground signal propagation between drones and ground nodes can be obstructed by obstacles such as tall buildings, potentially creating non-line-of-sight paths that degrade signal quality at the receiver and disrupt normal communication. Emerging smart reflective surface technology can address this challenge by proactively reshaping the propagation environment. Drones can fly close to ground terminals and communicate with them via line-of-sight links, thereby increasing air-to-ground data rates. Since line-of-sight channels dominate in air-to-ground systems, this increases the likelihood of eavesdroppers maliciously stealing user information.

[0004] Existing research has proposed maximizing the average security of a single-user communication system by jointly optimizing the UAV's horizontal trajectory, transmit beamforming, and smart reflector phase shift. Others, from a security perspective, consider the worst-case scenario caused by eavesdroppers and jointly optimize the UAV's trajectory, power control, and smart reflector phase shift to achieve the maximum security for a single user. Still others have proposed introducing smart reflectors as relays into UAV wireless information and power synchronization (SWIPT) networks to ensure the power receiver's energy requirements while improving system performance. However, with the increasing network capacity, single-user environments are no longer adequate for current conditions, increasing the risk of eavesdroppers maliciously stealing user information. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a secure communication method for drones with the assistance of a smart reflective surface, which maximizes the average confidentiality rate of the system and avoids the eavesdropping of user information by jointly optimizing user scheduling, three-dimensional trajectory of drones and reflection phase shift of smart reflective surface while ensuring the quality of legitimate user services.

[0006] The technical solution adopted in this invention is a secure communication method for unmanned aerial vehicles (UAVs) assisted by a smart reflective surface, which includes the following steps:

[0007] S1. Construct a downlink communication system consisting of a drone, a smart reflective surface, M legitimate users, and an active eavesdropper; the smart reflective surface is fixedly deployed on a building, and the drone communicates with the associated legitimate users through a link in each time slot, the link being either a direct link between the drone and the legitimate user or an indirect link between the drone and the smart reflective surface and the legitimate user.

[0008] S2. Set the target problem P1, which is to maximize the average security rate of the system by jointly optimizing user scheduling, intelligent reflective surface reflection phase shift and UAV three-dimensional trajectory.

[0009] S3. Arbitrarily set the reflection phase shift of the intelligent reflective surface and input the three-dimensional trajectory of the UAV into the target problem P1 set in step S2. The optimal UAV user scheduling is obtained by iterative optimization of the target problem P1.

[0010] S4. Input the optimal UAV user scheduling obtained in step S3 and the three-dimensional trajectory of the UAV arbitrarily set in step S3 into the target problem P1 set in step S2, and perform iterative optimization of the target problem P1 to obtain the optimal reflection phase shift of the intelligent reflective surface.

[0011] S5. Input the optimal UAV user scheduling obtained in step S3 and the optimal reflection phase shift of the intelligent reflective surface obtained in step S4 into the target problem P1 set in step S2, and perform iterative optimization of the target problem P1 to obtain the optimal three-dimensional trajectory of the UAV.

[0012] Preferably, in step S2, the target problem P1 is represented as:

[0013] P1:

[0014]

[0015]

[0016]

[0017] C4:z[1]=zinit ,z[T]=z final

[0018]

[0019]

[0020] Where A represents user scheduling.

[0021] α m [t] represents the scheduling variable representing the association between the drone and the legitimate user m within time slot t, α m [t]∈{0,1},α m [t] = 1 indicates that the drone serves the legitimate user m within time slot t. Otherwise α m [t] = 0; Φ represents the reflection phase shift vector of the intelligent reflective surface. Represents the nth time slot on the intelligent reflective surface within time slot t. x ,n y The phase shift of each reflecting element relative to the legitimate user m; Q represents the horizontal trajectory vector of the UAV. q[t]=[x[t],y[t]] T H represents the horizontal trajectory of the drone; H represents the vertical trajectory vector of the drone. H[t] represents the vertical trajectory of the UAV, and the three-dimensional trajectory of the UAV consists of a horizontal trajectory and a vertical trajectory; R m [t] represents the reachability and rate of ground user m within time slot t. Let M be the set of M legal users. m=0 indicates the index of the eavesdropper. This represents the total number of ground users, consisting of M legitimate users and one eavesdropper. R represents m,0 The upper bound of [t], R m,0 [t] represents the reachability and rate of the eavesdropper m=0 within time slot t; C1 represents the constraint on user association scheduling, C2 represents ensuring that the UAV associates with at most one legitimate user and transmits information within each time slot, C3 represents guaranteeing the quality of service for each legitimate user, C4 represents the initial and final position constraints of the UAV, C5 represents the constraint on the phase shift of the intelligent reflective surface, and C6 and C7 represent the trajectory constraints of the UAV; Q max Q represents the maximum horizontal movement distance allowed for a drone in each time slot. max =V h τ, H max H represents the maximum vertical movement distance allowed for a drone in each time slot.max =V z τ, where the length of each time slot is τ seconds, and the maximum horizontal and vertical flight speeds of the UAV within each time slot are V, respectively. h and V z .

[0022] Preferably, step S3 includes the following steps:

[0023] S3.1 Arbitrarily set the reflection phase shift of the intelligent reflective surface and input the three-dimensional trajectory of the UAV into the target problem P1 set in step S2. The target problem P1 is then iteratively optimized and transformed into target problem P2.

[0024] P2:

[0025] stC1,C2,C3; where... This indicates the system's average security level.

[0026] S3.2. Perform binary relaxation on the target problem P2 to transform it into the target problem. Regarding the target problem The optimal drone user scheduling is obtained through iterative optimization.

[0027] Preferably, in step S4, the optimal UAV user scheduling obtained in step S3 and the arbitrarily set three-dimensional trajectory of the UAV in step S3 are input into the target problem P1 set in step S2. The optimal reflection phase shift of the intelligent reflective surface at the legitimate user m within time slot t is obtained by iterative optimization of the target problem P1:

[0028] in,

[0029] This indicates the nth time interval on the intelligent reflective surface within time slot t. x ,n y The optimal phase shift generated by the reflecting elements relative to user m; It is a random scattering component that follows a circularly symmetric complex Gaussian distribution with zero mean and unit variance; θ (1) [t] and ζ (1) [t] represents the vertical and horizontal AOAs from the UAV to the smart reflective surface within time slot t, respectively; and These are represented as vertical AODs and horizontal AODs from the smart reflective surface to the user, respectively.

[0030] Preferably, step S5 includes the following steps:

[0031] Step 1: Optimize the horizontal trajectory of the drone:

[0032] S5.1. Input the optimal UAV user scheduling obtained in step S3, the optimal reflection phase shift of the smart reflective surface obtained in step S4, and the vertical trajectory of any given UAV into the target problem P1 set in step S2. The target problem P1 is then iteratively optimized and transformed into a non-convex target problem P3 for the horizontal trajectory of the UAV.

[0033] S5.2, the reachability and rate R of legal user m within time slot t. m [t] is used for iterative optimization: by

[0034] Where B represents the system bandwidth, P represents the UAV's transmit power, and σ 2 It is the power of Gaussian white noise, Ψ m [t] represents the phase shift matrix of the UAV-smart reflective surface-legal user link. in,

[0035] N x and N y These represent the number of reflective elements in the x and y directions of the intelligent reflective surface, respectively; towards R m Introducing slack variables in [t] and η Q [t] to replace g respectively m [t] and d (1) [t], and satisfy the constraints: and η Q [t]≥(d (1) [t]) 2 The reachable sum rate R of legal user m within time slot t is obtained. m [t] is redefined as:

[0036] in, right A first-order Taylor expansion yields lower bound The It is a concave function. in,

[0037] S5.3, for R m,0 upper bound of [t] Perform iterative optimization: by Introducing slack variable υ Q [t] and Replace (d0[t]) respectively. 2 and (d) (1) [t]) 2 and respectively satisfy constraint υ Q [t]≤(d0[t]) 2 and The reachable sum rate of the eavesdropper within time slot t Redefined as: The right-hand side of the newly introduced inequality constraint is convex with respect to q[t]. A first-order Taylor expansion at the l-th iteration transforms it into a concave function.

[0038] (d0[t]) 2 ≥q lb1 [t] = (d0[t]) (l) ) 2 +2(q[t] (l) -u0) T (q[t]-q[t] (l) ),

[0039] (d (1) [t]) 2 ≥q lb2 [t]=(d (1) [t] (l) ) 2 +2(q[t] (l) -q I ) T (q[t]-q[t] (l) );

[0040] S5.4. Based on the concave function obtained from steps S5.2 and S5.3, transform the non-convex target problem P3 of the UAV horizontal trajectory obtained in step S5.1 into a convex problem P4: in, Iterative optimization of the convex problem P4 yields the optimal horizontal trajectory for the UAV.

[0041] Step 2: Optimize the vertical trajectory of the drone:

[0042] S5.5. Input the optimal UAV user scheduling obtained in step S3, the optimal reflection phase shift of the smart reflective surface obtained in step S4, and the vertical trajectory of any given UAV into the target problem P1 set in step S2. The target problem P1 is then iteratively optimized and transformed into the optimization problem P5 of the UAV vertical trajectory. The optimization problem P5 is a non-convex optimization problem.

[0043] S5.6. Perform a convex optimization transformation on optimization problem P5 to obtain the vertical trajectory optimization problem P6 for the UAV: ​​P6:

[0044] stC4,C7

[0045]

[0046]

[0047]

[0048]

[0049] in,

[0050] and η H [t] are respectively (d m [t]) 2 and (d) m [t]) 2 The upper bound H obtained by performing a first-order Taylor expansion on q[t] at the l-th iteration is H. lb1 [t] and H lb2 [t] represents υ H [t] and The upper bound of q[t] is obtained by performing a first-order Taylor expansion at the l-th iteration; the vertical trajectory optimization problem P6 of the UAV is iteratively optimized to obtain the optimal vertical trajectory of the UAV.

[0051] The beneficial effects of this invention are as follows: The above-mentioned secure communication method for unmanned aerial vehicles (UAVs) assisted by an intelligent reflective surface employs an alternating optimization and continuous convex approximation algorithm. By jointly optimizing user scheduling, the reflection phase shift of the intelligent reflective surface, and the three-dimensional trajectory of the UAV, this method maximizes the average confidentiality rate of the system while ensuring the quality of service for legitimate users, thus preventing user information from being eavesdropped on. Attached Figure Description

[0052] Figure 1 This is a schematic diagram of the downlink communication system constructed according to the present invention;

[0053] Figure 2 This is a schematic diagram of the horizontal direction of the information transmission angle between the drone and the user in this invention;

[0054] Figure 3 This is a schematic diagram of the vertical direction of the information transmission angle between the drone and the user in this invention;

[0055] Figure 4 This is a schematic diagram showing the relationship between the average security level and the number of reflective elements achievable by the system assisted by the intelligent reflective surface in this invention;

[0056] Figure 5 This is a performance comparison chart obtained by using the method of this invention under four transmission schemes;

[0057] Figure 6 This is a comparison diagram of the security of the intelligent reflective surface under different phase control strategies in this invention;

[0058] Figure 7 This is a schematic diagram comparing the relationship between the system's average security level and the UAV's transmit power under different phase control strategies in this invention;

[0059] Figure 8 This is a comparative diagram showing the impact of the path loss index from the intelligent reflective surface to the ground user channel on the system's security level in this invention. Detailed Implementation

[0060] The invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can implement it based on the description. The scope of protection of the invention is not limited to these specific embodiments.

[0061] This invention relates to a secure communication method for unmanned aerial vehicles (UAVs) assisted by a smart reflective surface, the method comprising the following steps:

[0062] S1, such as Figure 1 As shown, a downlink communication system is constructed consisting of a drone, a smart reflective surface, M legitimate users, and an active eavesdropper. The smart reflective surface is fixedly deployed on a building. The drone communicates with the associated legitimate users through a link in each time slot. The link is either a direct link between the drone and the legitimate user or an indirect link between the drone, the smart reflective surface, and the legitimate user.

[0063] exist Figure 1 In this context, z[t] = [q[t]] T H[t]] represents the three-dimensional trajectory of the UAV within time slot t, where, H[t] and H[t] represent the horizontal and vertical trajectories of the UAV, respectively; for ease of description, Legitimate users and eavesdroppers are collectively referred to as ground users. in Let m represent the set of legitimate users, where m = 0 represents the index of the eavesdropper; the horizontal position of ground user m is represented as... The intelligent reflective surface is fixedly deployed on the surface of the building, and its horizontal position is indicated as follows: Vertical position is represented by H I The flight period of the UAV is set to T, the length of each time slot is τ seconds, and the maximum horizontal and vertical flight speeds of the UAV within each time slot are V and V, respectively. h and V zFor each legitimate user The reflection phase shift of the smart reflector in time slot t is expressed as:

[0064] exist Figure 1 In this invention, since line-of-sight channels dominate air-to-ground communication, the method considers not only the indirect link caused by passive reflection from the smart reflector but also the direct link from the UAV to the ground user. Within time slot t, the direct channel from the UAV to ground user m is represented as follows: Where β0 is the channel gain at a reference distance d = 1m; d m [t] is the distance from the UAV to the ground user m within time slot t, specifically expressed as... Within time slot t, the channel from the UAV to the smart reflector is represented as:

[0065] in,

[0066] and These represent the phase shifts of the UAV relative to the origin of the intelligent reflector in the x and y dimensions, respectively, within time slot t; d (1) [t] represents the distance from the UAV to the intelligent reflector within time slot t, specifically expressed as... like Figure 2 As shown, the relationship between the drone and the smart reflective surface in terms of angle and position can be used... and To describe, θ (1) [t] and ζ (1) [t] represents the vertical and horizontal AOAs from the UAV to the smart reflective surface within time slot t, respectively;

[0067] S2. Define the target problem P1, which is to maximize the average security rate of the system by jointly optimizing user scheduling, intelligent reflective surface reflection phase shift, and UAV three-dimensional trajectory. The specific process for defining the target problem P1 is as follows:

[0068] According to Shannon's formula, the legitimate users within time slot t are obtained. achievable rate:

[0069] Where, α m [t]∈{0,1} represents the interaction between the drone and the legitimate user within time slot t. Scheduling variables for associations; α m [t] = 1 indicates that the drone serves legitimate users within time slot t. Otherwise α m [t] = 0. B is the system bandwidth, P is the UAV's transmit power, and σ 2This is the power of Gaussian white noise. Ψ m [t] represents the phase shift matrix of the UAV-intelligent reflector-legitimate user link, specifically expressed as: in,

[0070]

[0071] Since small-scale fading between the smart reflector and the eavesdropper is difficult to obtain, R is obtained by solving Jensen's inequality. m,0 upper bound of [t] That is, considering the worst-case scenario where the eavesdropper m=0, the reachability rate is:

[0072] Among them, Ψ m,0 [t] is the phase shift matrix of the UAV-smart reflector-eavesdropper link, specifically represented as follows:

[0073] By jointly optimizing user scheduling Phase shift of intelligent reflective surface Horizontal trajectory of drone Vertical trajectory of the drone The following optimization problem is established: Among them, C1 is the constraint on user-associated scheduling; C2 ensures that in each time slot, the UAV is associated with at most one legitimate user and transmits information; C3 guarantees the quality of service for each legitimate user; C4 is the initial and final position constraint of the UAV; C5 is the constraint on the phase shift of the intelligent reflector; C6 and C7 are the trajectory constraints of the UAV, where Q... max =V h τ is the maximum horizontal movement distance allowed for a UAV in each time slot, H max =V z τ is the maximum vertical movement distance allowed for a drone in each time slot;

[0074] S3. Arbitrarily set the reflection phase shift of the intelligent reflective surface and input the three-dimensional trajectory of the UAV into the target problem P1 set in step S2. The optimal UAV user scheduling is obtained by iterative optimization of the target problem P1. The specific process is as follows:

[0075] S3.1 Arbitrarily set the reflection phase shift of the intelligent reflective surface and input the three-dimensional trajectory of the UAV into the target problem P1 set in step S2. The target problem P1 is then iteratively optimized and transformed into target problem P2.

[0076]

[0077] stC1,C2,C3; where... This indicates the system's average security level.

[0078] S3.2 Since constraint C1 is a binary constraint, problem P2 remains a non-convex problem. Therefore, binary relaxation of the target problem P2 transforms it into the target problem. Regarding the target problem The optimal drone user scheduling is obtained through iterative optimization; problem It is a standard convex optimization problem, which can be effectively solved using existing optimization tools such as CVX

[68] .

[0079] S4. Input the optimal UAV user scheduling obtained in step S3 and the 3D trajectory of the UAV arbitrarily set in step S3 into the target problem P1 set in step S2, and perform iterative optimization of the target problem P1 to obtain the optimal reflection phase shift of the intelligent reflective surface: in,

[0080] This indicates the nth time interval on the intelligent reflective surface within time slot t. x ,n y The optimal phase shift generated by the reflecting elements relative to user m; It is a random scattering component that follows a circularly symmetric complex Gaussian distribution with zero mean and unit variance; θ (1) [t] and ζ (1) [t] represents the vertical and horizontal AOAs from the UAV to the smart reflective surface within time slot t, respectively; and These are represented as vertical AODs and horizontal AODs from the smart reflective surface to the user, respectively.

[0081] S5. Input the optimal UAV user scheduling obtained in step S3 and the optimal reflection phase shift of the intelligent reflective surface obtained in step S4 into the target problem P1 set in step S2, and use the target problem P1 to iteratively optimize and obtain the optimal three-dimensional trajectory of the UAV; the specific process is as follows:

[0082] Step 1: Optimize the horizontal trajectory of the drone:

[0083] S5.1. Input the optimal UAV user scheduling obtained in step S3, the optimal reflection phase shift of the smart reflective surface obtained in step S4, and the vertical trajectory of any given UAV into the target problem P1 set in step S2. The target problem P1 is then iteratively optimized and transformed into a non-convex target problem P3 for the horizontal trajectory of the UAV. Due to the objective function Since constraint C3 with respect to the horizontal trajectory q[t] of the UAV is neither convex nor concave, problem P3 remains a non-convex problem, thus requiring adjustments to R. m [t] and Solve the non-convexity problem by optimizing separately;

[0084] S5.2, the reachability and rate R of legal user m within time slot t. m [t] is used for iterative optimization: by

[0085] Where B represents the system bandwidth, P represents the UAV's transmit power, and σ 2 It is the power of Gaussian white noise, Ψ m [t] represents the phase shift matrix of the UAV-smart reflective surface-legal user link. in,

[0086] N x and N y These represent the number of reflective elements in the x and y directions of the intelligent reflective surface, respectively; R in the formula m [t] is nonconvex with respect to q[t], so towards R m Introducing slack variables in [t] and η Q [t] to replace g respectively m [t] and d (1) [t], and satisfy the constraints: and η Q [t]≥(d (1) [t]) 2 The reachable sum rate R of legal user m within time slot t is obtained. m [t] is redefined as:

[0087] in, right A first-order Taylor expansion yields lower bound The It is a concave function. in,

[0088] S5.3, for R m,0 upper bound of [t] Perform iterative optimization: by

[0089] in the formula Since q[t] is nonconvex, a slack variable υ is introduced. Q [t] and Replace (d0[t]) respectively. 2 and (d) (1) [t]) 2 and respectively satisfy constraint υQ [t]≤(d0[t]) 2 and The reachable sum rate of the eavesdropper within time slot t Redefined as: It's about υ Q [t] and The convex function is q[t], and the right side of the newly introduced inequality constraint is convex with respect to q[t]. At the l-th iteration, a first-order Taylor expansion is performed to transform it into a concave function: (d0[t]). 2 ≥q lb1 [t] = (d0[t]) (l) ) 2 +2(q[t] (l) -u0) T (q[t]-q[t] (l) ), (d( 1 )[t]) 2 ≥q lb2 [t]=(d( 1 )[t] (l) ) 2 +2(q[t] (l) -q I ) T (q[t]-q[t] (l) );

[0090] S5.4. Based on the concave function obtained from steps S5.2 and S5.3, transform the non-convex target problem P3 of the UAV horizontal trajectory obtained in step S5.1 into a convex problem P4: in, Iterative optimization of the convex problem P4 yields the optimal horizontal trajectory for the UAV.

[0091] Step 2: Optimize the vertical trajectory of the drone:

[0092] S5.5. Input the optimal UAV user scheduling obtained in step S3, the optimal reflection phase shift of the smart reflective surface obtained in step S4, and the vertical trajectory of any given UAV into the target problem P1 set in step S2. The target problem P1 is then iteratively optimized and transformed into the optimization problem P5 of the UAV vertical trajectory. because Since constraint C3 is neither convex nor nonconvex, problem P5 remains a nonconvex optimization problem.

[0093] S5.6, due to R m [t] and The same transformation idea as in problem P3 can be used for convex optimization. Therefore, by performing a convex optimization transformation on optimization problem P5, we obtain the UAV vertical trajectory optimization problem P6:

[0094] P6:

[0095] stC4,C7

[0096]

[0097]

[0098]

[0099]

[0100] in,

[0101] and η H [t] are respectively (d m [t]) 2 and (d) m [t]) 2 The upper bound H obtained by performing a first-order Taylor expansion on q[t] at the l-th iteration is H. lb1 [t] and H lb2 [t] represents υ H [t] and The upper bound of q[t] is obtained by performing a first-order Taylor expansion at the l-th iteration; the vertical trajectory optimization problem P6 of the UAV is iteratively optimized to obtain the optimal vertical trajectory of the UAV.

[0102] In this invention, problem P1 is decomposed into three subproblems, which are then solved alternately using an algorithm based on alternating iteration and continuous convex approximation. Specifically, in the l-th iteration, the process of solving the four subproblems is as follows: the first step is to solve the subproblems. By considering the reflection phase shift of a given smart reflective surface and the three-dimensional trajectory of the UAV {Ψ (l) Q (l) H (l) Solve for the optimal user schedule A. (l+1) Used for the next iteration; the second step solves subproblem P4, given the optimal user scheduling, the optimal intelligent reflective surface reflection phase shift, and the vertical trajectory of any UAV {A}. (l+1) ,Ψ (l) H (l) The optimal horizontal trajectory Q of the UAV is obtained by iteratively solving using a continuous convex approximation algorithm. (l+1)The third step is to solve subproblem P6, given the optimal user scheduling, the optimal reflection phase shift of the intelligent reflective surface, and the horizontal trajectory of any UAV {A}. (l+1) ,Ψ (l+1) Q (l+1) The optimal vertical trajectory H of the UAV is obtained by iteratively solving using a continuous convex approximation algorithm. (l+1) Finally, continue iteratively updating the reflection phase shift of the smart reflective surface, repeating the above process until convergence, i.e., the threshold is less than ε3.

[0103] To verify the feasibility and effectiveness of the method of the present invention, a simulation experiment was conducted, as follows:

[0104] The horizontal coordinates of the legitimate user and the eavesdropper are set to [200, 100; 270, 50; 230, 60] respectively. T The coordinates of the intelligent reflective surface are [100,40], [200,0,10], and the initial and final positions of the UAV flight trajectory are set to [0,80,80] and [450,70,80], respectively. The maximum horizontal and vertical flight speeds of the UAV in each time slot are V. h =40m / s and V z =3m / s, the UAV's transmit power is P = 1W, the number of time slots is T = 50s, the length of each time slot is τ = 1s, and the noise power is σ. 2 = -80dBm, channel gain at reference distance is β0 = -20dB, path loss coefficient is η = 2.3, threshold ε1 = ε2 = 10 -4 and ε3 = 10 -8 .

[0105] The results of the simulation experiment are as follows Figures 4 to 8 As shown:

[0106] Figure 4 The iterative process of the algorithm proposed in this invention is demonstrated; the results show that the system assisted by the intelligent reflective surface can achieve a higher average security rate, and the number of reflective elements is directly proportional to the average security rate.

[0107] See Figure 5 The performance of four schemes was demonstrated: (1) the design proposed by the method of this invention; (2) the design without UAV trajectory optimization; (3) the design without user scheduling; and (4) the transmission design without intelligent reflective surface. The results show that for any transmission design assisted by intelligent reflective surface, the average security rate will increase with the increase of the number of reflective elements of intelligent reflective surface. The design proposed by this invention has the best performance among the four transmission designs.

[0108] See Figure 6This study demonstrates the security performance of intelligent reflective surfaces (IRS) under different phase control strategies. The results show that when the intelligent reflective surface employs optimal or random phase control, the average security rate of legitimate users increases with the increase of the number of reflective elements, and optimal phase control can achieve better security benefits than random phase control. In addition, compared with the transmission scheme without intelligent reflective surfaces, the scheme with optimal phase control of the reflective surface achieves significant performance gains.

[0109] See Figure 7 The study demonstrates the relationship between the average security level of the system and the UAV's transmit power under different phase control strategies. The results show that the system security level increases with the increase of the UAV's transmit power in all three cases. The security level with the assistance of intelligent reflective surfaces is much better than that without intelligent reflective surfaces, which fully reflects the benefits of intelligent reflective surfaces for secure network communication. In addition, the security level with the intelligent reflective surface adopting the optimal phase control strategy is better than that with the random phase control strategy.

[0110] See Figure 8 The study investigated the impact of the path loss index of the smart reflector-to-ground user channel on the system's security. When the path loss index increased from 2.1 to 2.9, the security of the system with smart reflector assistance deteriorated. This is because the increased path loss index leads to greater propagation loss in the wireless link, resulting in a decrease in the overall system security. However, the security with a smart reflector is still much better than that without a smart reflector, and the security level of the smart reflector using the optimal phase control strategy is better than that using the random phase control strategy.

[0111] Specifically, with the assistance of intelligent reflective surfaces, this invention provides a security design for UAV communication systems and proposes an algorithm based on alternating optimization and continuous convex approximation. Simulation results show that the algorithm achieves a better average system security rate than other benchmark schemes, demonstrating the significant advantages of intelligent reflective surfaces in secure transmission design.

[0112] This invention, while ensuring the quality of service for legitimate users, jointly optimizes user scheduling, UAV 3D trajectory, and reflection phase shift of intelligent reflective surfaces to maximize the system's average security rate. This problem is a mixed-integer non-convex optimization problem. Through alternating optimization and continuous convex approximation algorithms, a better suboptimal solution is found, thereby satisfying user service quality, improving system security, and combating eavesdropping.

[0113] This invention leverages the advantages of intelligent reflective surfaces to establish a dual-hop link, generating two communication modes for unmanned aerial vehicles (UAVs): direct communication and indirect communication. By jointly optimizing the UAV's three-dimensional trajectory, user scheduling, and the phase shift of the intelligent reflective surface, the system's flexibility is improved, covering a wider user base and achieving secure and efficient information transmission.

Claims

1. A secure communication method for unmanned aerial vehicles (UAVs) assisted by a smart reflective surface, characterized in that: The method includes the following steps: S1. Construct a downlink communication system consisting of a drone, a smart reflective surface, M legitimate users, and an active eavesdropper; the smart reflective surface is fixedly deployed on a building, and the drone communicates with the associated legitimate users through a link in each time slot, the link being either a direct link between the drone and the legitimate user or an indirect link between the drone and the smart reflective surface and the legitimate user. S2. Set the target problem P1, which is to maximize the average security rate of the system by jointly optimizing user scheduling, intelligent reflective surface reflection phase shift and UAV three-dimensional trajectory. S3. Arbitrarily set the reflection phase shift of the intelligent reflective surface and input the three-dimensional trajectory of the UAV into the target problem P1 set in step S2. The optimal UAV user scheduling is obtained by iterative optimization of the target problem P1. S4. Input the optimal UAV user scheduling obtained in step S3 and the three-dimensional trajectory of the UAV arbitrarily set in step S3 into the target problem P1 set in step S2, and perform iterative optimization of the target problem P1 to obtain the optimal reflection phase shift of the intelligent reflective surface. S5. Input the optimal UAV user scheduling obtained in step S3 and the optimal reflection phase shift of the intelligent reflective surface obtained in step S4 into the target problem P1 set in step S2, and perform iterative optimization of the target problem P1 to obtain the optimal three-dimensional trajectory of the UAV.

2. The method for secure communication of unmanned aerial vehicles (UAVs) assisted by a smart reflective surface according to claim 1, characterized in that: In step S2, the target problem P1 is represented as: Where A represents user scheduling. α m [t] represents the scheduling variable representing the association between the drone and the legitimate user m within time slot t, α m [t]∈{0,1},α m [t] = 1 indicates that the drone serves the legitimate user m within time slot t. Otherwise α m [t] = 0; Φ represents the reflection phase shift vector of the intelligent reflective surface. Represents the nth time slot on the intelligent reflective surface within time slot t. x ,n y The phase shift of each reflecting element relative to the legitimate user m; Q represents the horizontal trajectory vector of the UAV. q[t]=[x[t],y[t]] T H represents the horizontal trajectory of the drone; H represents the vertical trajectory vector of the drone. H[t] represents the vertical trajectory of the UAV, and the three-dimensional trajectory of the UAV consists of a horizontal trajectory and a vertical trajectory; R m [t] represents the reachability and rate of ground user m within time slot t. Let M be the set of M legal users. m=0 indicates the index of the eavesdropper. This represents the total number of ground users, consisting of M legitimate users and one eavesdropper. R represents m,0 The upper bound of [t], R m,0 [t] represents the reachability and rate of the eavesdropper m=0 within time slot t; C1 represents the constraint on user association scheduling, C2 represents ensuring that the UAV associates with at most one legitimate user and transmits information within each time slot, C3 represents guaranteeing the quality of service for each legitimate user, C4 represents the initial and final position constraints of the UAV, C5 represents the constraint on the phase shift of the intelligent reflective surface, and C6 and C7 represent the trajectory constraints of the UAV; Q max Q represents the maximum horizontal movement distance allowed for a drone in each time slot. max =V h τ, H max H represents the maximum vertical movement distance allowed for a drone in each time slot. max =V z τ, where the length of each time slot is τ seconds, and the maximum horizontal and vertical flight speeds of the UAV within each time slot are V, respectively. h and V z .

3. The method for secure communication of unmanned aerial vehicles (UAVs) assisted by a smart reflective surface according to claim 2, characterized in that: Step S3 includes the following steps: S3.1 Arbitrarily set the reflection phase shift of the intelligent reflective surface and input the three-dimensional trajectory of the UAV into the target problem P1 set in step S2. The target problem P1 is then iteratively optimized and transformed into target problem P2. in, This indicates the system's average security level. S3.

2. Perform binary relaxation on the target problem P2 to transform it into the target problem. For the target problem The optimal drone user scheduling is obtained through iterative optimization.

4. The method for secure communication of unmanned aerial vehicles (UAVs) assisted by a smart reflective surface according to claim 3, characterized in that: In step S4, the optimal UAV user scheduling obtained in step S3 and the arbitrarily set 3D trajectory of the UAV in step S3 are input into the target problem P1 set in step S2. The optimal reflection phase shift of the intelligent reflective surface at the legitimate user m within time slot t is obtained by iterative optimization of the target problem P1: in, This indicates the nth time interval on the intelligent reflective surface within time slot t. x ,n y The optimal phase shift generated by the reflecting elements relative to user m; It is a random scattering component that follows a circularly symmetric complex Gaussian distribution with zero mean and unit variance; θ (1) [t] and ζ (1) [t] represents the vertical and horizontal AOAs from the UAV to the smart reflective surface within time slot t, respectively; and These are represented as vertical AODs and horizontal AODs from the smart reflective surface to the user, respectively.

5. The method for secure communication of unmanned aerial vehicles (UAVs) assisted by a smart reflective surface according to claim 4, characterized in that: Step S5 includes the following steps: Step 1: Optimize the horizontal trajectory of the drone: S5.

1. Input the optimal UAV user scheduling obtained in step S3, the optimal reflection phase shift of the smart reflective surface obtained in step S4, and the vertical trajectory of any given UAV into the target problem P1 set in step S2. The target problem P1 is then iteratively optimized and transformed into a non-convex target problem P3 for the horizontal trajectory of the UAV. S5.2, the reachability and rate R of legal user m within time slot t. m [t] is used for iterative optimization: by Where B represents the system bandwidth, P represents the UAV's transmit power, and σ 2 It is the power of Gaussian white noise, Ψ m [t] represents the phase shift matrix of the UAV-smart reflective surface-legal user link. in, N x and N y These represent the number of reflective elements in the x and y directions of the intelligent reflective surface, respectively; towards R m Introducing slack variables in [t] and η Q [t] to replace g respectively m [t] and d (1) [t], and satisfy the constraints: and η Q [t]≥(d (1) [t]) 2 The reachable sum rate R of legal user m within time slot t is obtained. m [t] is redefined as: in, right A first-order Taylor expansion yields lower bound The It is a concave function. in, S5.3, for R m,0 upper bound of [t] Perform iterative optimization: by Introducing slack variable υ Q [t] and Replace (d0[t]) respectively. 2 and (d) (1) [t]) 2 and respectively satisfy constraint υ Q [t]≤(d0[t]) 2 and The reachable sum rate of the eavesdropper within time slot t Redefined as: The right-hand side of the newly introduced inequality constraint is convex with respect to q[t]. A first-order Taylor expansion at the l-th iteration transforms it into a concave function. (d0[t]) 2 ≥q lb1 [t]=(d0[t] (l) ) 2 +2(q[t] (l) -u0) T (q[t]-q[t] (l) ), (d (1) [t]) 2 ≥q lb2 [t]=(d (1) [t] (l) ) 2 +2(q[t] (l) -q I ) T (q[t]-q[t] (l) ); S5.

4. Based on the concave function obtained from steps S5.2 and S5.3, transform the non-convex target problem P3 of the UAV horizontal trajectory obtained in step S5.1 into a convex problem P4: in, Iterative optimization of the convex problem P4 yields the optimal horizontal trajectory for the UAV. Step 2: Optimize the vertical trajectory of the drone: S5.

5. Input the optimal UAV user scheduling obtained in step S3, the optimal reflection phase shift of the smart reflective surface obtained in step S4, and the vertical trajectory of any given UAV into the target problem P1 set in step S2. The target problem P1 is then iteratively optimized and transformed into the optimization problem P5 of the UAV vertical trajectory. The optimization problem P5 is a non-convex optimization problem. S5.

6. Perform a convex optimization transformation on optimization problem P5 to obtain the vertical trajectory optimization problem P6 for the UAV: in, and η H [t] are respectively (d m [t]) 2 and (d) m [t]) 2 The upper bound H obtained by performing a first-order Taylor expansion on q[t] at the l-th iteration is H. lb1 [t] and H lb2 [t] represents υ H [t] and The upper bound of q[t] is obtained by performing a first-order Taylor expansion at the l-th iteration; the vertical trajectory optimization problem P6 of the UAV is iteratively optimized to obtain the optimal vertical trajectory of the UAV.

Images