Robust unmanned aerial vehicle terahertz safety communication method

Abstract

The invention discloses a robust unmanned aerial vehicle terahertz safety communication method. The method comprises the steps of firstly establishing an unmanned aerial vehicle terahertz safety communication system; establishing an optimization problem combining the unmanned aerial vehicle trajectory Q, the artificial noise vector f and the transmitted beam forming vector w by taking the minimization of the total transmitted power of the unmanned aerial vehicle flight base station as a target; and finally, a block coordinate descent method and an S process positive semi-definite relaxation successive convex approximation algorithm are utilized to convert and solve the optimization problem, and a corresponding optimal unmanned aerial vehicle trajectory set Qb, an optimal artificial noise vector fb and an optimal transmission beam forming vector wb are obtained when the total transmission power P of the unmanned aerial vehicle flight base station is minimum. The communication method is accurate in channel estimation, so that the energy consumption of the unmanned aerial vehicle base station is low.

Description

Technical Field

[0001] The present invention relates to the field of communications, and more specifically, to a robust unmanned aerial vehicle terahertz secure communication method. Background Art

[0002] As a potential 6G technology, terahertz is expected to solve the problems of spectrum scarcity and limited capacity in current wireless systems. However, terahertz communication is affected by severe propagation loss and molecular absorption, which limits the transmission distance and communication capacity. In addition, obstacles in the propagation environment may directly block communication. However, drones are suitable for using terahertz channel communication for information transmission due to their air-to-ground link characteristics.

[0003] Traditional UAV terahertz communication methods assume that the channel state is too ideal, and therefore often fail to meet the needs of real environments. In reality, there are eavesdroppers, which leads to imperfect channel states. At the same time, in reality, UAVs are affected by environmental speeds such as wind, which causes the fuselage to be unstable, resulting in channel estimation errors for both users and eavesdroppers, and the UAV consumes a huge amount of energy to compensate for this error.

[0004] The prior art discloses a downlink channel estimation method in UAV communication. This method estimates the complete channel state information based on the channel frequency response of the pilot signal and the pilot position using the learned long short-term memory network. This method does not take into account the actual situation of the channel state. Summary of the invention

[0005] The present invention aims to solve the problem of errors in channel estimation in the prior art, which leads to high energy consumption, and provides a robust terahertz secure communication method for unmanned aerial vehicles.

[0006] The primary purpose of the present invention is to solve the above technical problems. The technical solution of the present invention is as follows:

[0007] A robust UAV terahertz secure communication method, comprising:

[0008] S1: Establish a UAV terahertz secure communication system, which includes a UAV flight base station, multiple users and an eavesdropper;

[0009] S2: Based on the UAV terahertz secure communication system, the joint UAV trajectory is established with the goal of minimizing the total transmission power of the UAV flight base station. Optimization problem of artificial noise vector f and transmit beamforming vector w;

[0010] S3: Using the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm, the optimization problem is transformed and solved to obtain the optimal drone trajectory set corresponding to the minimum total transmission power P of the drone flight base station. Optimal artificial noise vector f b and the optimal transmit beamforming vector w b .

[0011] Furthermore, in the drone terahertz secure communication system, the drone base station communicates with each user via a first channel, and the drone base station communicates with the eavesdropper via a second channel;

[0012] The estimated first channel uncertainty gain for the nth time slot is as follows:

[0013]

[0014] The terahertz path gain of the first channel is as follows:

[0015]

[0016] The transmission rate of the first channel of the kth user in the nth time slot is as follows:

[0017]

[0018] The eavesdropper's eavesdropping rate on the first channel of the kth user in the nth time slot is as follows:

[0019]

[0020] Among them, j and k represent user numbers, n represents the time slot number, represents the terahertz path gain of the first channel, represents the communication frequency, C represents the speed of light, d k [n] represents the distance from the kth user to the drone base station in the nth time slot, k f represents the medium absorption factor; represents the conjugate transposed matrix of the first channel uncertainty gain of the kth user in the nth time slot, represents the conjugate transposed matrix of the second channel uncertainty gain of the eavesdropper in the nth time slot; w k [n] is the transmit beamforming vector of the kth user in the nth time slot, f[n] is the artificial noise vector of the nth time slot, σ e represents the power spectral density of the second channel white noise; σ k represents the white noise power spectrum density of the first channel of the kth user, Represents the guidance vector from the drone base station to the kth user.

[0021] Furthermore, the optimization problem is as follows:

[0022]

[0023] The constraints are:

[0024] (x[n+1]-x[n]) 2 +(y[n+1]-y[n]) 2 ≤(v max ·ΔT) 2

[0025]

[0026] q[0]=q 0

[0027] k represents the user number, k represents all users; n represents the time slot number, N represents all time slots; w k [n] represents the beamforming vector of the kth user in the nth time slot; w k represents the beamforming vector of the kth user, w k ={w k [1],w k [2],…,w k [N]}; w represents the user's beamforming vector w = {w 1 ,w 2 ,…,w K}; f[n] represents the artificial noise vector sent by the UAV base station in the nth time slot; f[n] represents the artificial noise vector sent by the UAV base station in the nth time slot, f represents the artificial noise vector sent by the UAV base station, f={f[1],f[2],…,f[N]}; represents the trajectory set of the UAV; x[n] represents the horizontal coordinate of the UAV in the nth time slot, y[n] represents the vertical coordinate of the UAV in the nth time slot; v max represents the maximum speed of the drone; ΔT represents the duration of each time slot; Δh k [n] represents the first channel gain estimation error of the kth user in the nth time slot, Δh e [n] represents the second channel gain estimation error of the nth time slot; ε k [n] represents the upper limit of the error of the first channel gain estimation of the kth user in the nth time slot, ε e [n] represents the upper limit of the second channel gain estimation error of the nth time slot; R k [n] represents the transmission rate of the first channel of the kth user in the nth time slot, represents the eavesdropping rate of the eavesdropper on the first channel of the kth user in the nth time slot; represents the minimum transmission rate of the first channel of the kth user, Indicates the maximum transmission rate of the second channel; q[0], q 0 represents the initial trajectory of the drone,

[0028] Further, in step S3, the optimization problem is transformed and solved by using the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm, including:

[0029] S301: Based on the joint UAV trajectory The optimization problem of the artificial noise vector f and the transmit beamforming vector w is derived to obtain the first optimization target and the second optimization target;

[0030] S302: Use the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm to solve the first optimization objective and the second optimization objective; obtain the optimal drone trajectory set corresponding to the minimum total transmission power P of the drone flight base station Optimal artificial noise vector f b and the optimal transmit beamforming vector w b .

[0031] Furthermore, the first optimization objective is as follows:

[0032]

[0033] The constraints are:

[0034] W k [n]≥0,F[n]≥0

[0035]

[0036] rank(W k [n])=1,rank(F[n])=1

[0037] k represents the user number, K represents all users; n represents the time slot number, N represents all time slots; w k [n] represents the beamforming vector of the kth user in the nth time slot, W k [n] represents the beamforming vector product of the kth user in the nth time slot, w k represents the beamforming vector of the kth user, w k ={w k [1],w k [2],…,w k [N]}; w represents the user's beamforming vector w = {w 1 ,w 2 ,…,w K}; f[n] represents the artificial noise vector sent by the UAV base station in the nth time slot, and F[n] represents the product of the artificial noise vector sent by the UAV base station in the nth time slot; λ k[n] represents the non-negative slack variable of the kth user in the nth time slot, I M represents the identity matrix; Γ k [n] represents the gamma function of the kth user in the nth time slot; represents the conjugate transposed matrix of the estimated first channel uncertainty gain of the kth user in the nth time slot, represents the estimated first channel uncertainty gain of the kth user in the nth time slot; e [n] represents the second channel slack variable of the nth time slot, W e [n] represents the eavesdropper beamforming vector product of the nth time slot, represents the conjugate transposed matrix of the estimated second channel uncertainty gain for the nth time slot, represents the estimated second channel uncertainty gain of the nth time slot; Γ e [n] represents the gamma function of the eavesdropper at the nth time slot.

[0038] Furthermore, the second optimization objective is as follows:

[0039]

[0040] The constraints are:

[0041] (x[n+1]-x[n]) 2 +(y[n+1]-y[n]) 2 ≤(v max ·ΔT) 2

[0042]

[0043] l represents the current iteration number; 2 represents the trajectory set of the drone; k represents the user number, n represents the time slot number; x[n] represents the horizontal coordinate of the nth time slot of the drone, x[n] (l-1) represents the horizontal coordinate of the nth time slot of the UAV in the l-1th iteration, y[n] represents the vertical coordinate of the nth time slot of the UAV, y[n] (l-1) represents the ordinate of the nth time slot of the UAV in the l-1th iteration, (x k ,y k ,0) represents the position of the kth user, (x e ,y e ) represents the location of the eavesdropper; v max represents the maximum speed of the drone; ΔT represents the duration of each time slot; β 1,k [n] represents the first intermediate value of the kth user in the nth time slot; β 2,k [n] represents the second median value of the kth user's nth time slot; β3,k [n] represents the third median value of the kth user's nth time slot; represents the square of the slack variable of the kth user in the nth time slot, u k [n] represents the slack variable of the kth user in the nth time slot, represents the square of the slack variable of the kth user in the nth time slot at the l-1th iteration, u k [n] (l-1) represents the slack variable of the k-th user in the n-th time slot at the l-1-th iteration; represents the minimum transmission rate of the first channel of the kth user; k f represents the medium absorption factor, H represents the height of the drone; β 1,e [n] represents the first intermediate value of the eavesdropper's nth time slot, β 2,e [n] represents the second median value of the eavesdropper’s nth time slot, β 3,e [n] represents the third median value of the eavesdropper’s nth time slot; Indicates the maximum transmission rate of the second channel; u e [n] represents the slack variable of the eavesdropper in the nth time slot, represents the square of the slack variable of the eavesdropper in the nth time slot, u e [n] (l-1) represents the slack variable of the nth time slot of the eavesdropper in the l-1th iteration, represents the square of the slack variable of the nth time slot of the eavesdropper in the l-1th iteration.

[0044] Furthermore, the first intermediate value β of the nth time slot of the kth user 1,k [n] are as follows:

[0045]

[0046] The second intermediate value β of the kth user's nth time slot 2,k [n] are as follows:

[0047]

[0048] The third intermediate value β of the kth user's nth time slot 3,k [n] are as follows:

[0049]

[0050] Among them, ρ 1 represents the first error probability; j, k represent the user number, n represents the time slot number; l represents the current iteration number, represents the trajectory variable of the kth user in the nth time slot at the l-1th iteration, w k[n] represents the beamforming vector of the kth user in the nth time slot, and f[n] represents the artificial noise vector sent by the drone base station in the nth time slot.

[0051] Furthermore, the first intermediate value β of the nth time slot of the eavesdropper 1,e [n] are as follows:

[0052]

[0053] The second intermediate value β of the eavesdropper's nth time slot 2,e [n] are as follows:

[0054]

[0055] The third intermediate value β of the eavesdropper in the nth time slot 3,e [n] are as follows:

[0056]

[0057] Among them, ρ 2 represents the second error probability; j and k represent the user number, n represents the time slot number; l represents the current iteration number, represents the trajectory variable of the nth time slot of the eavesdropper in the l-1th iteration, w k [n] represents the beamforming vector of the kth user in the nth time slot, and f[n] represents the artificial noise vector sent by the drone base station in the nth time slot.

[0058] Furthermore, the first error probability ρ 1 as follows:

[0059]

[0060] The second error probability ρ 2 as follows:

[0061]

[0062] Where f represents the artificial noise vector sent by the drone base station, C represents the speed of light, and σ e represents the power spectral density of the second channel white noise; σ k represents the white noise power spectral density of the first channel of the kth user.

[0063] Furthermore, in step S302, the first optimization objective and the second optimization objective are solved by using the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm, including:

[0064] S30201: Initialize the trajectory set of the first UAV The first artificial noise vector f1 and the first user beamforming vector w 1 , and calculate the first total transmission power P 1 ;

[0065] S30202: Based on the trajectory set of the first UAV Solve the first optimization objective and obtain the second artificial noise vector f 2 and the second user beamforming vector w 2 ;

[0066] S30203: According to the second artificial noise vector f 2 and the second user beamforming vector w 2 , solve the second optimization objective and obtain the trajectory set of the second UAV

[0067] S30204: According to the second artificial noise vector f 2 , the second user beamforming vector w 2 , the trajectory set of the second UAV Calculate the second total transmission power P 2 ;

[0068] S30205: Judgment Is it greater than the preset value? If so, the trajectory set of the second UAV As a new first drone track collection Execute step S30202; if not, set the trajectory of the second drone As the optimal UAV trajectory set The second artificial noise vector f 2 As the optimal artificial noise vector f b , the second user beamforming vector w 2 As the optimal transmit beamforming vector w b .

[0069] Compared with the prior art, the present invention has the following beneficial effects:

[0070] The present invention establishes an optimization problem by establishing a UAV terahertz secure communication system and taking minimizing the total transmission power of the UAV flight base station as the goal; finally, the optimization problem is solved to make the UAV flight base station channel estimation more accurate, thereby reducing the energy consumption of the UAV flight base station. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 A flowchart of a robust UAV terahertz secure communication method is provided in an embodiment.

[0072] Figure 2The flowchart of the embodiment provided is to transform and solve the optimization problem by using the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm.

[0073] Figure 3 A trajectory diagram of the drone flight provided in the embodiment.

[0074] Figure 4 The embodiment provides a convergence line graph of the optimization problem solved by transforming the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm. DETAILED DESCRIPTION

[0075] The drawings are for illustrative purposes only and should not be construed as limiting the present patent;

[0076] In order to better illustrate the present embodiment, some parts in the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product;

[0077] It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0078] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0079] Example

[0080] like Figure 1 As shown, a robust UAV terahertz secure communication method comprises:

[0081] S1: Establish a UAV terahertz secure communication system, which includes a UAV flight base station, multiple users and an eavesdropper;

[0082] S2: Based on the UAV terahertz secure communication system, the joint UAV trajectory is established with the goal of minimizing the total transmission power of the UAV flight base station. Optimization problem of artificial noise vector f and transmit beamforming vector w;

[0083] S3: Using the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm, the optimization problem is transformed and solved to obtain the optimal drone trajectory set corresponding to the minimum total transmission power P of the drone flight base station. Optimal artificial noise vector f b and the optimal transmit beamforming vector w b .

[0084] Furthermore, in the drone terahertz secure communication system, the drone base station communicates with each user via a first channel, and the drone base station communicates with the eavesdropper via a second channel;

[0085] The estimated first channel uncertainty gain for the nth time slot is as follows:

[0086]

[0087] The terahertz path gain of the first channel is as follows:

[0088]

[0089] The transmission rate of the first channel of the kth user in the nth time slot is as follows:

[0090]

[0091] The eavesdropper's eavesdropping rate on the first channel of the kth user in the nth time slot is as follows:

[0092]

[0093] Among them, j and k represent user numbers, n represents the time slot number, represents the terahertz path gain of the first channel, represents the communication frequency, C represents the speed of light, d k [n] represents the distance from the kth user to the drone base station in the nth time slot, k f represents the medium absorption factor; represents the conjugate transposed matrix of the first channel uncertainty gain of the kth user in the nth time slot, represents the conjugate transposed matrix of the second channel uncertainty gain of the eavesdropper in the nth time slot; w k [n] is the transmit beamforming vector of the kth user in the nth time slot, f[n] is the artificial noise vector of the nth time slot, σ e represents the power spectral density of the second channel white noise; σ k represents the white noise power spectrum density of the first channel of the kth user, Represents the guidance vector from the drone base station to the kth user.

[0094] In a specific embodiment, the steering vector from the drone base station to the kth user is as follows:

[0095]

[0096] The estimated second channel uncertainty gain for the nth time slot is given by:

[0097]

[0098] The terahertz path gain of the second channel is as follows:

[0099]

[0100] The guidance vector from the drone base station to the eavesdropper is as follows:

[0101]

[0102] sinφ k [n] represents the arrival angle of the kth user to the drone base station at the nth time slot, d and λ are constants, and M is the total number of users. e [n] represents the distance from the eavesdropper to the drone base station, represents the terahertz path gain of the second channel, sinφ e [n] represents the arrival angle from the eavesdropper to the drone base station in the nth time slot.

[0103] Furthermore, the optimization problem is as follows:

[0104]

[0105] The constraints are:

[0106] (x[n+1]-x[n]) 2 +(y[n+1]-y[n]) 2 ≤(v max ·ΔT) 2

[0107]

[0108] q[0]=q 0

[0109] k represents the user number, K represents all users; n represents the time slot number, N represents all time slots; w k [n] represents the beamforming vector of the kth user in the nth time slot; w k represents the beamforming vector of the kth user, w k ={w k [1],w k [2],…,w k [N]}; w represents the user's beamforming vector w = {w 1 ,w 2 ,…,w K}; f[n] represents the artificial noise vector sent by the UAV base station in the nth time slot; f[n] represents the artificial noise vector sent by the UAV base station in the nth time slot, f represents the artificial noise vector sent by the UAV base station, f={f[1],f[2],…,f[N]}; 2 represents the trajectory set of the UAV; x[n] represents the horizontal coordinate of the UAV in the nth time slot, y[n] represents the vertical coordinate of the UAV in the nth time slot; v max represents the maximum speed of the drone; ΔT represents the duration of each time slot; Δh k[n] represents the first channel gain estimation error of the kth user in the nth time slot, Δh e [n] represents the second channel gain estimation error of the nth time slot; ε k [n] represents the upper limit of the error of the first channel gain estimation of the kth user in the nth time slot, ε e [n] represents the upper limit of the second channel gain estimation error of the nth time slot; R k [n] represents the transmission rate of the first channel of the kth user in the nth time slot, represents the eavesdropping rate of the eavesdropper on the first channel of the kth user in the nth time slot; represents the minimum transmission rate of the first channel of the kth user, Indicates the maximum transmission rate of the second channel; q[0], q 0 represents the initial trajectory of the drone,

[0110] In a specific embodiment, the first channel gain estimation error Δh of the kth user in the nth time slot is k [n], estimated first channel uncertainty gain of the kth user in the nth time slot The first channel uncertainty gain h of the kth user in the nth time slot k [n], the upper limit of the error of the first channel gain estimation of the kth user in the nth time slot ε k [n]Satisfy:

[0111]

[0112] In a specific embodiment, the second channel gain estimation error Δh of the nth time slot is e [n], estimated second channel uncertainty gain for the nth time slot The second channel uncertainty gain h of the eavesdropper in the nth time slot e [n], the upper limit of the error of the second channel gain estimation in the nth time slot ε e [n]Satisfy:

[0113]

[0114] In a specific embodiment, the signal sent by the drone is as follows:

[0115]

[0116] The signal received by the kth user in the nth time slot is as follows:

[0117]

[0118] The signal received by the eavesdropper in the nth time slot is as follows:

[0119]

[0120] n k [n] represents the Gaussian white noise received by the kth user in the nth time slot, n e [n] represents the Gaussian white noise received by the eavesdropper in the nth time slot, and s[n] represents a constant.

[0121] In one specific embodiment, the constraint

[0122]

[0123] Can be written as:

[0124]

[0125] It can be further deduced as:

[0126]

[0127] In a specific embodiment, the optimization problem can be further transformed into:

[0128]

[0129] The constraints are:

[0130]

[0131] rank(W k [n])=1,rank(F[n])=1

[0132] W k [n]≥0,F[n]≥0

[0133] In a specific embodiment, q[n] = (x[n], y[n], H), and the position of the drone is the position of the drone base station.

[0134] Furthermore, if Figure 2 As shown, in step S3, the optimization problem is transformed and solved by using the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm, including:

[0135] S301: Based on the joint UAV trajectory The optimization problem of the artificial noise vector f and the transmit beamforming vector w is derived to obtain the first optimization target and the second optimization target;

[0136] S302: Use the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm to solve the first optimization objective and the second optimization objective; obtain the optimal drone trajectory set corresponding to the minimum total transmission power P of the drone flight base station Optimal artificial noise vector f b and the optimal transmit beamforming vector w b .

[0137] Furthermore, the first optimization objective is as follows:

[0138]

[0139] The constraints are:

[0140] W k [n]≥0,F[n]≥0

[0141]

[0142] rank(w k [n])=1,rank(F[n])=1

[0143] k represents the user number, K represents all users; n represents the time slot number, n represents all time slots; w k [n] represents the beamforming vector of the kth user in the nth time slot, W k [n] represents the beamforming vector product of the kth user in the nth time slot, w k represents the beamforming vector of the kth user, w k ={w k [1],w k [2],…,w k [N]}; w represents the user's beamforming vector w = {w 1 ,w 2 ,…,w K}; f[n] represents the artificial noise vector sent by the UAV base station in the nth time slot, and F[n] represents the product of the artificial noise vector sent by the UAV base station in the nth time slot; λ k [n] represents the non-negative slack variable of the kth user in the nth time slot, I M represents the identity matrix; Γ k [n] represents the gamma function of the kth user in the nth time slot; represents the conjugate transposed matrix of the estimated first channel uncertainty gain of the kth user in the nth time slot, represents the estimated first channel uncertainty gain of the kth user in the nth time slot; e [n] represents the second channel slack variable of the nth time slot, W e [n] represents the eavesdropper beamforming vector product of the nth time slot, represents the conjugate transposed matrix of the estimated second channel uncertainty gain for the nth time slot, represents the estimated second channel uncertainty gain of the nth time slot; Γ e [n] represents the gamma function of the eavesdropper at the nth time slot.

[0144] Furthermore, the second optimization objective is as follows:

[0145]

[0146] The constraints are:

[0147] (x[n+1]-x[n]) 2 +(y[n+1]-y[n]) 2 ≤(v max ·ΔT) 2

[0148]

[0149] l represents the current iteration number; represents the trajectory set of the drone; k represents the user number, n represents the time slot number; x[n] represents the horizontal coordinate of the nth time slot of the drone, x[n] (l-1) represents the horizontal coordinate of the nth time slot of the UAV in the l-1th iteration, y[n] represents the vertical coordinate of the nth time slot of the UAV, y[n] (l-1) represents the ordinate of the nth time slot of the UAV in the l-1th iteration, (x k ,y k ,0) represents the position of the kth user, (x e ,y e ) represents the location of the eavesdropper; v max represents the maximum speed of the drone; ΔT represents the duration of each time slot; β 1,k [n] represents the first intermediate value of the kth user in the nth time slot; β 2,k [n] represents the second median value of the kth user's nth time slot; β 3,k [n] represents the third median value of the kth user's nth time slot; represents the square of the slack variable of the kth user in the nth time slot, u k [n] represents the slack variable of the kth user in the nth time slot, represents the square of the slack variable of the kth user in the nth time slot at the l-1th iteration, u k [n] (l-1) represents the slack variable of the k-th user in the n-th time slot at the l-1-th iteration; represents the minimum transmission rate of the first channel of the kth user; k f represents the medium absorption factor, H represents the height of the drone; β 1,e[n] represents the first intermediate value of the eavesdropper's nth time slot, β 2,e [n] represents the second median value of the eavesdropper’s nth time slot, β 3,e [n] represents the third median value of the eavesdropper’s nth time slot; Indicates the maximum transmission rate of the second channel; u e [n] represents the slack variable of the eavesdropper in the nth time slot, represents the square of the slack variable of the eavesdropper in the nth time slot, u e [n] (l-1) represents the slack variable of the nth time slot of the eavesdropper in the l-1th iteration, represents the square of the slack variable of the nth time slot of the eavesdropper in the l-1th iteration.

[0150] Furthermore, the first intermediate value β of the nth time slot of the kth user 1,k [n] are as follows:

[0151]

[0152] The second intermediate value β of the kth user's nth time slot 2,k [n] are as follows:

[0153]

[0154] The third intermediate value β of the kth user's nth time slot 3,k [n] are as follows:

[0155]

[0156] Among them, ρ 1 represents the first error probability; j, k represent the user number, n represents the time slot number; l represents the current iteration number, represents the trajectory variable of the kth user in the nth time slot at the l-1th iteration, w k [n] represents the beamforming vector of the kth user in the nth time slot, and f[n] represents the artificial noise vector sent by the drone base station in the nth time slot.

[0157] Furthermore, the first intermediate value β of the nth time slot of the eavesdropper 1,e [n] are as follows:

[0158]

[0159] The second intermediate value β of the eavesdropper's nth time slot 2,e [n] are as follows:

[0160]

[0161] The third intermediate value β of the eavesdropper in the nth time slot 3,e [n] are as follows:

[0162]

[0163] Among them, ρ 2 represents the second error probability; j and k represent the user number, n represents the time slot number; l represents the current iteration number, represents the trajectory variable of the nth time slot of the eavesdropper in the l-1th iteration, w k [n] represents the beamforming vector of the kth user in the nth time slot, and f[b] represents the artificial noise vector sent by the drone base station in the nth time slot.

[0164] Furthermore, the first error probability ρ 1 as follows:

[0165]

[0166] The second error probability ρ 2 as follows:

[0167]

[0168] Where f represents the artificial noise vector sent by the drone base station, C represents the speed of light, and σ e represents the power spectral density of the second channel white noise; σ k represents the white noise power spectral density of the first channel of the kth user.

[0169] Furthermore, in step S302, the first optimization objective and the second optimization objective are solved by using the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm, including:

[0170] S30201: Initialize the trajectory set of the first UAV The first artificial noise vector f 1 and the first user beamforming vector w 1 , and calculate the first total transmission power P 1 ;

[0171] S30202: Based on the trajectory set of the first UAV Solve the first optimization objective and obtain the second artificial noise vector f 2 and the second user beamforming vector w 2 ;

[0172] S30203: According to the second artificial noise vector f 2 and the second user beamforming vector w 2, solve the second optimization objective and obtain the trajectory set of the second UAV

[0173] S30204: According to the second artificial noise vector f 2 , the second user beamforming vector w 2 , the trajectory set of the second UAV Calculate the second total transmission power P 2 ;

[0174] S30205: Judgment Is it greater than the preset value? If so, the trajectory set of the second UAV As a new first drone track collection Execute step S30202; if not, set the trajectory of the second drone As the optimal UAV trajectory set The second artificial noise vector f 2 As the optimal artificial noise vector f b , the second user beamforming vector w 2 As the optimal transmit beamforming vector w b .

[0175] In a specific embodiment, the total number of users K is set to 2, the UAV flight time T is set to 20s, the UAV flight altitude H is set to 15m, and the UAV maximum speed v is set to max is 4m / s; Figure 3 As shown in the figure, the drone flies from the starting point (0m, 0m, 15m) to the end point (0m, 30m, 15m). The positions of the users are (6m, 5m) and (6m, 25m), respectively, and the position of the eavesdropper is (-2m, 15m). The channel normalization uncertainty upper limit is Minimum transmission rate of the first channel of the kth user Maximum transmission rate of the second channel Dielectric absorption factor k f =0.05, communication frequency is 0.8THZ; the square of the white noise power spectrum density of the second channel σ e 2 , the square of the white noise power spectrum density of the first channel of the kth user σ k 2 , both are -110dbm.

[0176] like Figure 4 As shown in the figure, the algorithm can converge under different safety rate thresholds with good convergence. In addition, as the user's minimum safety rate threshold increases, the total transmission power required by the base station increases, and the base station needs a larger transmission power to meet higher security performance.

[0177] The same or similar reference numerals correspond to the same or similar components;

[0178] The terms used in the drawings to describe positional relationships are only used for illustrative purposes and should not be construed as limiting this patent;

[0179] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A robust UAV terahertz secure communication method, characterized in that: include: S1: Establish a UAV terahertz secure communication system, which includes a UAV flight base station, multiple users and an eavesdropper; S2: Based on the UAV terahertz secure communication system, the joint UAV trajectory is established with the goal of minimizing the total transmission power of the UAV flight base station. Optimization problem of artificial noise vector f and transmit beamforming vector w; S3: Using the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm, the optimization problem is transformed and solved to obtain the optimal drone trajectory set corresponding to the minimum total transmission power P of the drone flight base station. Optimal artificial noise vector f b and the optimal transmit beamforming vector w b .

2. A robust UAV terahertz secure communication method according to claim 1, characterized in that: In the drone terahertz secure communication system, the drone base station communicates with each user via a first channel, and the drone base station communicates with the eavesdropper via a second channel; The estimated first channel uncertainty gain for the nth time slot is as follows: The terahertz path gain of the first channel is as follows: The transmission rate of the first channel of the kth user in the nth time slot is as follows: The eavesdropper's eavesdropping rate on the first channel of the kth user in the nth time slot is as follows: Among them, j and k represent user numbers, n represents the time slot number, represents the terahertz path gain of the first channel, represents the communication frequency, C represents the speed of light, d k [n] represents the distance from the kth user to the drone base station in the nth time slot, k f represents the medium absorption factor; represents the conjugate transposed matrix of the first channel uncertainty gain of the kth user in the nth time slot, represents the conjugate transposed matrix of the second channel uncertainty gain of the eavesdropper in the nth time slot; w k [n] is the transmit beamforming vector of the kth user in the nth time slot, f[n] is the artificial noise vector of the nth time slot, σ e represents the power spectral density of the second channel white noise; σ k represents the white noise power spectrum density of the first channel of the kth user, Represents the guidance vector from the drone base station to the kth user.

3. A robust UAV terahertz secure communication method according to claim 2, characterized in that: The optimization problem is as follows: The constraints are: (x[n+1]-x[n]) 2 +(y[n+1]-y[n]) 2 ≤(v max ·ΔT) 2 q[0]=q0 k represents the user number, K represents all users; n represents the time slot number, N represents all time slots; w k [n] represents the beamforming vector of the kth user in the nth time slot; w k represents the beamforming vector of the kth user, w k ={w k [1],w k [2],…,w k [N]}; w represents the user’s beamforming vector w = {w1,w2,…,w K }; f[n] represents the artificial noise vector sent by the UAV base station in the nth time slot; f[n] represents the artificial noise vector sent by the UAV base station in the nth time slot, f represents the artificial noise vector sent by the UAV base station, f={f[1],f[2],…,f[N]}; Q represents the trajectory set of the UAV; x[n] represents the horizontal coordinate of the UAV in the nth time slot, y[n] represents the vertical coordinate of the UAV in the nth time slot; v max represents the maximum speed of the drone; Δt represents the duration of each time slot; Δh k [n] represents the first channel gain estimation error of the kth user in the nth time slot, Δh e [n] represents the second channel gain estimation error of the nth time slot; ε k [n] represents the upper limit of the error of the first channel gain estimation of the kth user in the nth time slot, ε e [n] represents the upper limit of the second channel gain estimation error of the nth time slot; R k [n] represents the transmission rate of the first channel of the kth user in the nth time slot, represents the eavesdropping rate of the eavesdropper on the first channel of the kth user in the nth time slot; represents the minimum transmission rate of the first channel of the kth user, represents the maximum transmission rate of the second channel; q[0], q0 represent the initial trajectory of the UAV, q={q[0],q[1],…,q[N]}.

4. A robust UAV terahertz secure communication method according to claim 2, characterized in that: In step S3, the optimization problem is transformed and solved by using the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm, including: S301: Based on the joint UAV trajectory The optimization problem of the artificial noise vector f and the transmit beamforming vector w is derived to obtain the first optimization target and the second optimization target; S302: Use the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm to solve the first optimization objective and the second optimization objective; obtain the optimal drone trajectory set corresponding to the minimum total transmission power P of the drone flight base station Optimal artificial noise vector f b and the optimal transmit beamforming vector w b .

5. A robust UAV terahertz secure communication method according to claim 4, characterized in that: The first optimization goal is as follows: The constraints are: W k [n]≥0,F[n]≥0 rank(W k [n])=1,rank(F[n])=1 k represents the user number, K represents all users; n represents the time slot number, N represents all time slots; w k [n] represents the beamforming vector of the kth user in the nth time slot, W k [n] represents the beamforming vector product of the kth user in the nth time slot, w k represents the beamforming vector of the kth user, w k ={w k [1],w k [2],…,w k [N]}; w represents the user’s beamforming vector w = {w1,w2,…,w K }; f[n] represents the artificial noise vector sent by the UAV base station in the nth time slot, and F[n] represents the product of the artificial noise vector sent by the UAV base station in the nth time slot; λ k [n] represents the non-negative slack variable of the kth user in the nth time slot, I M represents the identity matrix; Γ k [n] represents the gamma function of the kth user in the nth time slot; represents the conjugate transposed matrix of the estimated first channel uncertainty gain of the kth user in the nth time slot, represents the estimated first channel uncertainty gain of the kth user in the nth time slot; e [n] represents the second channel slack variable of the nth time slot, W e [n] represents the eavesdropper beamforming vector product of the nth time slot, represents the conjugate transposed matrix of the estimated second channel uncertainty gain for the nth time slot, represents the estimated second channel uncertainty gain of the nth time slot; Γ e [n] represents the gamma function of the nth time slot of the eavesdropper.

6. A robust UAV terahertz secure communication method according to claim 4, characterized in that: The second optimization objective is as follows: The constraints are: (x[n+1]-x[n]) 2 +(y[n+1]-y[n]) 2 ≤(v max ·ΔT) 2 l represents the current iteration number; represents the trajectory set of the drone; k represents the user number, n represents the time slot number; x[n] represents the horizontal coordinate of the nth time slot of the drone, x[n] (l-1) represents the horizontal coordinate of the nth time slot of the UAV in the l-1th iteration, y[n] represents the vertical coordinate of the nth time slot of the UAV, y[n] (l-1) represents the ordinate of the nth time slot of the UAV in the l-1th iteration, (x k ,y k ,0) represents the position of the kth user, (x e ,y e ) represents the location of the eavesdropper; v max represents the maximum speed of the drone; ΔT represents the duration of each time slot; β 1,k [n] represents the first intermediate value of the kth user in the nth time slot; β 2,k [n] represents the second median value of the kth user's nth time slot; β 3,k [n] represents the third median value of the kth user's nth time slot; represents the square of the slack variable of the kth user in the nth time slot, u k [n] represents the slack variable of the kth user in the nth time slot, represents the square of the slack variable of the kth user in the nth time slot at the l-1th iteration, u k [n] (l-1) represents the slack variable of the k-th user in the n-th time slot at the l-1-th iteration; represents the minimum transmission rate of the first channel of the kth user; k f represents the medium absorption factor, H represents the height of the drone; β 1,e [n] represents the first intermediate value of the eavesdropper's nth time slot, β 2,e [n] represents the second median value of the eavesdropper’s nth time slot, β 3,e [n] represents the third median value of the eavesdropper’s nth time slot; Indicates the maximum transmission rate of the second channel; u e [n] represents the slack variable of the eavesdropper in the nth time slot, represents the square of the slack variable of the eavesdropper in the nth time slot, u e [n] (l-1) represents the slack variable of the nth time slot of the eavesdropper in the l-1th iteration, represents the square of the slack variable of the nth time slot of the eavesdropper in the l-1th iteration.

7. A robust UAV terahertz secure communication method according to claim 6, characterized in that: The first intermediate value β of the kth user in the nth time slot 1,k [n] are as follows: The second intermediate value β of the kth user's nth time slot 2,k [n] are as follows: The third intermediate value β of the kth user's nth time slot 3,k [n] are as follows: Where ρ1 represents the first error probability; j and k represent the user numbers, n represents the time slot number; l represents the current iteration number, represents the trajectory variable of the kth user in the nth time slot at the l-1th iteration, w k [n] represents the beamforming vector of the kth user in the nth time slot, and f[n] represents the artificial noise vector sent by the drone base station in the nth time slot.

8. A robust UAV terahertz secure communication method according to claim 7, characterized in that: The first intermediate value β of the eavesdropper's nth time slot 1,e [n] are as follows: The second intermediate value β of the eavesdropper's nth time slot 2,e [n] are as follows: The third intermediate value β of the eavesdropper in the nth time slot 3,e [n] are as follows: Where ρ2 represents the second error probability; j and k represent the user numbers, n represents the time slot number; l represents the current iteration number, represents the trajectory variable of the nth time slot of the eavesdropper in the l-1th iteration, w k [n] represents the beamforming vector of the kth user in the nth time slot, and f[n] represents the artificial noise vector sent by the drone base station in the nth time slot.

9. A robust UAV terahertz secure communication method according to claim 8, characterized in that: The first error probability ρ1 is as follows: The second error probability ρ2 is as follows: Where f represents the artificial noise vector sent by the drone base station, C represents the speed of light, and σ e represents the power spectral density of the second channel white noise; σ k represents the white noise power spectral density of the first channel of the kth user.

10. A robust UAV terahertz secure communication method according to claim 4, characterized in that: In step S302, the first optimization objective and the second optimization objective are solved by using the block coordinate descent method and the S-process semi-positive definite relaxation successive convex approximation algorithm, including: S30201: Initialize the trajectory set of the first UAV The first artificial noise vector f 1 and the first user beamforming vector w 1 , and calculate the first total transmission power P 1 ; S30202: Based on the trajectory set of the first UAV Solve the first optimization objective and obtain the second artificial noise vector f 2 and the second user beamforming vector w 2 ; S30203: According to the second artificial noise vector f 2 and the second user beamforming vector w 2 , solve the second optimization objective and obtain the trajectory set of the second UAV S30204: According to the second artificial noise vector f 2 , the second user beamforming vector w 2 , the trajectory set of the second UAV Calculate the second total transmission power P 2 ; S30205: Judgment Is it greater than the preset value? If so, the trajectory set of the second UAV As a new first drone track collection Execute step S30202; if not, set the trajectory of the second drone As the optimal UAV trajectory set The second artificial noise vector f 2 As the optimal artificial noise vector f b , the second user beamforming vector w 2 As the optimal transmit beamforming vector w b .

Images