Track optimization method for covert communication under assistance of unmanned aerial vehicle

By building a drone-assisted hidden communication system, the UAV trajectory and channel parameters are optimized using binary hypothesis test and approximate convex optimization methods, the problem of combining drone trajectory optimization and hidden communication under time-varying channels is solved, and communication is improved Quality and efficiency.

CN119028184APending Publication Date: 2024-11-26KUNMING UNIVERSITY
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410989224.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-23
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

The prior art fails to combine the trajectory optimization of drones with hidden communication under time-varying channel conditions, resulting in limited communication quality and efficiency.

Method used

By building a trajectory optimization system for hidden communication assisted by drone, using binary hypothesis test to calculate false alarm probability and missed detection probability, optimize the drone flight trajectory and transmission power, combine slack variables and approximate convex optimization methods to optimize the unmanned The machine's trajectory and channel parameters are used to achieve optimal communication.

Benefits of technology

It effectively improves the stability and coverage of hidden communication, solves the problems of narrow communication range and signal blocking, and realizes the combination of drone trajectory optimization and hidden communication under time-varying channels.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119028184A_ABST
    Figure CN119028184A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of wireless and mobile communication and covert communication, and discloses an unmanned aerial vehicle assisted covert communication trajectory optimization method, which constructs an unmanned aerial vehicle assisted covert communication trajectory optimization system, and comprises an unmanned aerial vehicle sending end, a plurality of legal ground user receiving ends and a monitor Warden. The system further comprises two channel links: a covert communication link from the unmanned aerial vehicle sending end to each ground user and an eavesdropping link from the unmanned aerial vehicle sending end to a monitor Warden. According to the method, the unmanned aerial vehicle is used for assisting covert communication, the track of the unmanned aerial vehicle is optimized, the problems that a traditional communication method is narrow in communication range, a base station is fixed and inflexible, and signals are blocked are effectively solved, the stability and coverage of covert communication are improved, variables are successively optimized by adopting an approximate convex optimization method, and the communication efficiency is improved. And the optimal variable is taken.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical fields of wireless and mobile communications and concealed communications, and in particular to a trajectory optimization method for concealed communications assisted by an unmanned aerial vehicle. Background Art

[0002] With the widespread application of the Internet of Things in the fields of smart medical care, smart home, and smart transportation, the demand for wireless systems is also huge due to the need to connect a large number of wireless devices. Due to its high flexibility and low cost, drones have become an important intermediary in modern networks to improve communication quality and efficiency. In addition, drones are also used for data collection in wireless sensor networks to expand the coverage of the Internet of Things in information centers. Since the communication between drones and ground users is usually unimpeded, the high-quality communication channel between drones and ground users is vulnerable to eavesdropping attacks.

[0003] Covert communication is a promising technique to achieve data privacy protection through wireless communication processes in drone-assisted wireless systems. These two key security techniques are usually applied in different scenarios and are usually studied independently. To date, only a few research works have considered these two techniques simultaneously.

[0004] It is worth noting that in the current urban communication environment, the use of drones for communication is ineffective, but there has been no attempt to combine drone trajectory optimization with covert communication under time-varying channel conditions. Summary of the invention

[0005] 1. Technical issues to be solved

[0006] In view of the shortcomings of the prior art, the present invention provides a trajectory optimization method for covert communication assisted by a drone, which has the advantages of achieving simultaneous covert communication and trajectory optimization of the drone under time-varying channels, thereby solving the above-mentioned technical problems.

[0007] (II) Technical solution

[0008] To achieve the above object, the present invention provides the following technical solution: a trajectory optimization method for covert communication assisted by a drone, comprising the following steps:

[0009] S1. Construct a trajectory optimization system for UAV-assisted covert communication, including a UAV transmitter, multiple legitimate ground user receivers, and a monitor Warden;

[0010] It also includes two channel links: a covert communication link from the UAV transmitter to each ground user and an eavesdropping link from the UAV transmitter to the listener Warden. The channel gains between the UAV transmitter and the kth ground user and the listener Warden are h dand h k , and the channel gain obeys the complex Gaussian distribution;

[0011] S2. Perform binary hypothesis test calculation on the listener Warden to obtain the false alarm probability P r1 and the probability of missed detection P r2 , and calculate the total detection error rate ξ of the listener Warden, and finally get the optimal average minimum error detection probability ξ of the listener Warden * ;

[0012] S3. Analyze the trajectory of the drone and combine it with the optimal average minimum error detection probability ξ * Optimize the flight trajectory of the drone and ultimately obtain the optimal transmission power of the drone;

[0013] S4. Propose an optimization problem. The specific expression is as follows:

[0014]

[0015] st

[0016] v[0]=v0,v[N]=v N

[0017] q[0]=q0,q[N]=q N

[0018] ||v[n]||≤v max ,n=1,…,N

[0019] ||v[n]||≥v min ,n=1,…,N

[0020] ||a[n]||≤a max ,n=1,…,N

[0021]

[0022] v[n]=v[n-1]+a[n]Δt,n=2,…,N

[0023] in, represents the objective function of optimization, It means maximizing the objective function under the premise that q is the independent variable, q represents the position of the drone, N represents the total number of time slots, and K represents the number of users. represents the variance of the noise uncertainty at the kth user, Δt represents the unit time slot, and h k[n] represents the channel gain at the kth user in the nth time slot, log2 represents the logarithmic function with base 2, q[0] represents the position of the drone at the starting point, q0 represents the initial starting position of the drone, q[N] represents the position of the drone at the end point, q N represents the initialization of the UAV end position, v[0] represents the speed of the UAV at the starting point, v0 represents the initialization of the UAV starting speed, v[N] represents the speed of the UAV at the end point, v N represents the initialization UAV termination speed, v[n] represents the UAV flight speed at the nth time slot, v[n-1] represents the UAV flight speed at the n-1th time slot, and v max and v min represents the maximum and minimum values ​​of the UAV's flight speed, a[n] represents the UAV's flight acceleration at the nth time slot, and a max represents the maximum value of the UAV's flight acceleration, st represents the constraint, and P * [n] represents the optimal transmission power of the UAV transmitter, ||v[n]|| and ||a[n]|| represent the Euclidean norm operation of the UAV’s velocity and acceleration in the nth time slot, respectively;

[0024] S5, solving the optimization problem proposed in step S4, specifically comprising the following steps:

[0025] S5.1. Introduce a slack variable constraint and transform the optimization objective function. The specific expression is as follows:

[0026] Slack variable constraints:

[0027] μ k [n]≥||q[n]-w k || α ,n=1,2,…,N

[0028] The objective function after conversion:

[0029]

[0030] Among them, ||*|| represents the Euclidean norm, α represents the path fading coefficient, q[n] represents the position of the UAV at the nth time slot, H represents the flight altitude of the UAV, and μ k [n] represents a slack variable introduced at the kth user in the nth time slot, w k represents the position of the kth user, β0 represents the channel gain per unit distance, represents the variance of the noise uncertainty at the kth user;

[0031] S5.2, perform first-order Taylor expansion on the objective function converted in step S5.1 to obtain the following expression:

[0032]

[0033] in, represents the optimized objective function, P * [n] represents the optimal transmission power of the UAV transmitter, H represents the flight altitude of the UAV, μ k [n] represents a slack variable introduced at the kth user in the nth time slot, where is the optimization variable, and μ k [n] is the iteration variable;

[0034] S5.3. Change the constraint of the UAV flight speed in step S5.1. The specific expression is as follows:

[0035]

[0036] Among them, ν * [n] represents the flight speed of the drone at the nth time slot, and this variable is the optimization variable, ||*|| 2 represents the square of the Euclidean norm, Indicates v * The transpose of v[n] represents the flight speed of the drone at the nth time slot, v min Indicates the minimum flight speed of the drone;

[0037] S5.4. Use the SCA method to solve the transformed optimization problem through iteration and approximation methods.

[0038] As a preferred technical solution of the present invention, step S2 comprises the following steps:

[0039] S2.1. Obtain the drone that Warden eavesdropped on in the nth time slot to send a covert signal y to the ground user d [n], its specific expression is as follows:

[0040]

[0041] Among them, x[n] represents the signal sent by the drone in the nth time slot, H0 represents the null hypothesis, H1 represents the alternative hypothesis, and n d [n] represents the additive Gaussian white noise interference signal received by the listener in the nth time slot, and its mean is 0 and the variance is The normal distribution of y d[n] indicates that the eavesdropper Warden eavesdrops on the hidden signal in the nth time slot, P[n] indicates the transmission power of the drone at this time, h d represents the channel gain between the drone transmitter and the listener, Follow uniform distribution;

[0042] S2.2, the power expression of the signal received by the listener Warden is as follows:

[0043]

[0044] Among them, P d [n] represents the power of the signal received by the listener Warden, h d [n] represents the channel fading coefficient between the transmitter of the drone and the listener Warden in the nth time slot, H0 represents the null hypothesis, H1 represents the alternative hypothesis, and P[n] represents the transmission power of the drone at this time;

[0045] S2.3. The listener Warden uses a binary hypothesis test based on the received signal power to determine whether the sender is communicating covertly with the ground user. D0 and D1 represent the listener's decision to determine whether the sender is communicating covertly with the ground user. D0 means that the listener Warden believes that the sender is not sending covert information, and D1 means that the listener Warden believes that the sender is communicating covertly with the ground user. The expression of the total detection error probability ξ[n] of the listener Warden in the nth time slot is as follows:

[0046] ξ[n]=P r0 [n]+P r1 [n]

[0047] Where ξ[n] represents the total detection error probability of the nth time slot listener Warden, P r1 [n] represents the false alarm probability of the nth time slot, P r2 [n] represents the missed detection probability of the nth time slot, and the expressions are as follows:

[0048]

[0049]

[0050] Among them, P r1 [n] represents the false alarm probability of the nth time slot, Express satisfaction The probability of Express satisfaction The probability, P de [n] represents the error detection threshold of the listener Warden at the nth time slot, hd [n] represents the channel fading coefficient between the transmitting end of the drone and the listener Warden in the nth time slot, and P[n] represents the transmission power between the drone and each ground user in the nth time slot;

[0051] S2.4. Calculate the optimal average minimum error detection probability ξ of the listener * [n], the specific expression is as follows:

[0052]

[0053] Among them, ξ * [n] represents the minimum total detection error probability of the listener in the nth time slot, h d [n] represents the channel fading coefficient between the drone transmitter and the listener Warden in the nth time slot, and P[n] represents the transmission power between the drone and each ground user in the nth time slot.

[0054] As the preferred technical solution of the present invention, the false alarm probability P in step S2.3 is r1 The expression is as follows:

[0055]

[0056] Among them, P r1 represents the false alarm probability of the monitor Warden, P de [n] represents the error detection threshold of the listener Warden at the nth time slot, σ 2 represents the noise uncertainty at Warden The distribution parameters obeyed, The obedience parameter is [σ 2 -ρ,σ 2 +ρ] uniform distribution, ρ represents a minimum value;

[0057] Missed detection probability P r2 The expression is as follows:

[0058]

[0059] Where P[n] represents the transmission power between the drone and each ground user in the nth time slot, Q[n] = h d [n]P[n],P de [n] represents the error detection threshold of the listener Warden at the nth time slot, σ 2 represents the noise uncertainty at Warden The distribution parameters obeyed, The obedience parameter is [σ 2 -ρ,σ 2 +ρ] uniform distribution, ρ represents a minimum value;

[0060] The specific expression of the total detection error probability ξ[n] of the nth time slot listener Warden is as follows:

[0061]

[0062] Among them, P de [n] represents the error detection threshold of the listener Warden at the nth time slot, σ 2 represents the noise uncertainty at Warden The distribution parameters obeyed, The obedience parameter is [σ 2 -ρ,σ 2 +ρ], ρ represents a minimum value, P[n] represents the transmission power of the UAV and the ground users in the nth time slot, h d [n] represents the channel fading coefficient between the UAV transmitter and the listener Warden in the nth time slot.

[0063] As a preferred technical solution of the present invention, the specific process of step S3 is as follows:

[0064] S3.1. Introduce a random variable C[n] that reflects the channel uncertainty between the drone and the listener in the actual scenario;

[0065] S3.2, based on the introduced random variable C[n] and combined with the optimal average minimum error detection probability ξ * [n], get the minimum average detection error probability

[0066] Minimum average detection error probability The expression is as follows:

[0067]

[0068] Among them, E C[n] (*) represents the expectation of C[n], ξ * (C[n]) represents the minimum detection error probability with C[n] as the independent variable, and At the same time E C[n] (1) = 1, H represents the flight altitude of the drone, represents the uncertainty value used to simulate the channel between the drone and the listener, and β0 represents the channel gain per unit distance;

[0069] S3.3, the random variable C[n] is subject to Gaussian distribution, that is, C[n]~N(λ[n]+2,4λ[n]+4), and the following inequality is combined with the minimum average detection error probability Performing the operation, inequality 1 is as follows:

[0070]

[0071] θ[n]=E{|C[n]-E(C[n])|}

[0072]

[0073]

[0074] Among them, θ[n], p1[n], λ[n] are parameters used to express their corresponding expressions, and ρ represents a very small value. And combine inequality 1 with the minimum average detection error probability After calculation, we get the following inequality 2:

[0075]

[0076] Among them, θ[n], p1[n], λ[n] are parameters used to express their corresponding expressions, C[n] represents the random variable of the nth time slot, H represents the flight altitude of the drone, β0 represents the channel gain per unit distance, ρ represents a very small value, and P[n] represents the transmission power of the drone and each ground user in the nth time slot. When p2(C[n])=ρ d When , the transmitting power of the UAV transmitter reaches the maximum value. A value representing the uncertainty of the channel between the simulated drone and the listener.

[0077] As a preferred technical solution of the present invention, the specific expression in step S3.1 is as follows:

[0078]

[0079] Among them, ||*|| represents the Euclidean norm, α represents the path fading coefficient, q[n] represents the position of the UAV at the nth time slot, and C[n] represents the random variable of the nth time slot. represents the uncertainty value used to simulate the channel between the drone and the listener, w d Indicates the location of the listener Warden.

[0080] As a preferred technical solution of the present invention, the optimal transmission power P of the UAV transmitter is * [n]The expression is as follows:

[0081]

[0082] in, P max Indicates the maximum transmission power of the drone, Represents the smallest of two elements.

[0083] Compared with the prior art, the present invention provides a trajectory optimization method for covert communication assisted by a drone, which has the following beneficial effects:

[0084] The present invention utilizes drones to assist in covert communications and optimizes their trajectories, thereby effectively solving the problems faced by traditional communication methods, such as narrow communication range, inflexible fixed base stations, and signal blocking, and improves the stability and coverage of covert communications. In addition, an approximate convex optimization method is used to successively optimize the variables until the optimal variables are taken. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 A schematic diagram of a cellular network covert eavesdropping system under channel uncertainty of the present invention;

[0086] Figure 2 Schematic diagram of drone trajectories in three situations of the cellular network covert eavesdropping method under channel uncertainty of the present invention;

[0087] Figure 3 Schematic diagram of the curve of total throughput changing with concealment constraints in three cases of the cellular network covert eavesdropping method under channel uncertainty of the present invention;

[0088] Figure 4 It is a schematic diagram showing the total throughput of the cellular network covert eavesdropping method under channel uncertainty of the present invention in three cases with the maximum transmission power of the UAV;

[0089] Figure 5 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0090] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0091] See also Figure 1-Figure 5 , a trajectory optimization method for UAV-assisted covert communication, comprising the following steps:

[0092] S1. Construct a trajectory optimization system for UAV-assisted covert communication, including a UAV transmitter, multiple legitimate ground user receivers, and a monitor Warden;

[0093] It also includes two channel links: a covert communication link from the UAV transmitter to each ground user and an eavesdropping link from the UAV transmitter to the listener Warden. The channel gains between the UAV transmitter and the kth ground user and the listener Warden are h d and h k , and the channel gain obeys the complex Gaussian distribution;

[0094] in, Follows uniform distribution [σ 2 -ρ,σ 2 +ρ], and ρ is a very small value

[0095] In the system, the subscripts d and k represent the listener Warden and the kth ground user GroundUser, respectively. The channel gain between the UAV transmitter and the kth ground user and the listener is h d and h k These channel gains all have zero mean and variance It is assumed that all channels follow Rayleigh fading and remain constant within a time slot but vary independently between time slots.

[0096] S2. Perform binary hypothesis test calculation on the listener Warden to obtain the false alarm probability P r1 and the probability of missed detection P r2 , and calculate the total detection error rate ξ of the listener Warden, and finally get the optimal average minimum error detection probability ξ of the listener Warden * ;

[0097] S2.1. Obtain the drone that Warden eavesdropped on in the nth time slot to send a covert signal y to the ground user d [n], its specific expression is as follows:

[0098]

[0099] Among them, x[n] represents the signal sent by the drone in the nth time slot, H0 represents the null hypothesis, H1 represents the alternative hypothesis, and n d [n] represents the additive Gaussian white noise interference signal received by the listener in the nth time slot, and its mean is 0 and the variance is The normal distribution of y d [n] indicates that the eavesdropper Warden eavesdrops on the hidden signal in the nth time slot, P[n] indicates the transmission power of the drone at this time, h d represents the channel gain between the drone transmitter and the listener, Follow uniform distribution;

[0100] S2.2, the power expression of the signal received by the listener Warden is as follows:

[0101]

[0102] Among them, P d [n] represents the power of the signal received by the listener Warden, h d [n] represents the channel fading coefficient between the transmitter of the drone and the listener Warden in the nth time slot, H0 represents the null hypothesis, H1 represents the alternative hypothesis, and P[n] represents the transmission power of the drone at this time;

[0103] S2.3. The listener Warden uses a binary hypothesis test based on the received signal power to determine whether the sender is communicating covertly with the ground user. D0 and D1 represent the listener's decision to determine whether the sender is communicating covertly with the ground user. D0 means that the listener Warden believes that the sender is not sending covert information, and D1 means that the listener Warden believes that the sender is communicating covertly with the ground user. The expression of the total detection error probability ξ[n] of the listener Warden in the nth time slot is as follows:

[0104] ξ[n]=P r0 [n]+P r1 [n]

[0105] Where ξ[n] represents the total detection error probability of the nth time slot listener Warden, P r0 [n] represents the false alarm probability of the nth time slot, P r1 [n] represents the missed detection probability of the nth time slot, and the expressions are as follows:

[0106]

[0107]

[0108] Among them, P r0 [n] represents the false alarm probability of the nth time slot, Express satisfaction The probability of Express satisfaction The probability, P de [n] represents the error detection threshold of the listener Warden at the nth time slot, h d [n] represents the channel fading coefficient between the transmitting end of the drone and the listener Warden in the nth time slot, and P[n] represents the transmission power between the drone and each ground user in the nth time slot;

[0109] S2.4. Calculate the optimal average minimum error detection probability ξ of the listener* [n], the specific expression is as follows:

[0110]

[0111] Among them, ξ * [n] represents the minimum total detection error probability of the listener in the nth time slot, h d [n] represents the channel fading coefficient between the drone transmitter and the listener Warden in the nth time slot, and P[n] represents the transmission power between the drone and each ground user in the nth time slot.

[0112] As the preferred technical solution of the present invention, the false alarm probability P in step S2.3 is r1 The expression is as follows:

[0113]

[0114] Among them, P r1 represents the false alarm probability of the monitor Warden, P de [n] represents the error detection threshold of the listener Warden at the nth time slot, σ 2 represents the noise uncertainty at Warden The distribution parameters obeyed, The obedience parameter is [σ 2 -ρ,σ 2 +ρ] uniform distribution, ρ represents a minimum value;

[0115] Missed detection probability P r2 The expression is as follows:

[0116]

[0117] Where P[n] represents the transmission power between the drone and each ground user in the nth time slot, Q[n] = h d [n]P[n],P de [n] represents the error detection threshold of the listener Warden at the nth time slot, σ 2 represents the noise uncertainty at Warden The distribution parameters obeyed, The obedience parameter is [σ 2 -ρ,σ 2 +ρ] uniform distribution, ρ represents a minimum value;

[0118] The specific expression of the total detection error probability ξ[n] of the nth time slot listener Warden is as follows:

[0119]

[0120] Among them, Pde [n] represents the error detection threshold of the listener Warden at the nth time slot, σ 2 represents the noise uncertainty at Warden The distribution parameters obeyed, The obedience parameter is [σ 2 -ρ,σ 2 +ρ], ρ represents a minimum value, P[n] represents the transmission power of the UAV and the ground users in the nth time slot, h d [n] represents the channel fading coefficient between the transmitter of the drone and the listener Warden in the nth time slot

[0121] S3. Analyze the trajectory of the drone and combine it with the optimal average minimum error detection probability ξ * Optimize the flight trajectory of the drone and ultimately obtain the optimal transmission power of the drone;

[0122] S3.1. Introduce a random variable C[n] that reflects the channel uncertainty between the drone and the listener in the actual scenario;

[0123] The specific expression in step S3.1 is as follows:

[0124]

[0125] Among them, ||*|| represents the Euclidean norm, α represents the path fading coefficient, q[n] represents the position of the UAV at the nth time slot, and C[n] represents the random variable of the nth time slot. represents the uncertainty value used to simulate the channel between the drone and the listener, w d Indicates the location of the listener Warden

[0126] S3.2, based on the introduced random variable C[n] and combined with the optimal average minimum error detection probability ξ * [n], get the minimum average detection error probability

[0127] Minimum average detection error probability The expression is as follows:

[0128]

[0129] Among them, E C[n] (*) represents the expectation of C[n], ξ * (C[n]) represents the minimum detection error probability with C[n] as the independent variable, and At the same time E C[n] (1) = 1, H represents the flight altitude of the drone, represents the uncertainty value used to simulate the channel between the drone and the listener, and β0 represents the channel gain per unit distance;

[0130] S3.3, the random variable C[n] is subject to Gaussian distribution, that is, C[n]~N[λ[n]+2,4λ[n]+4), and the following inequality is combined with the minimum average detection error probability Performing the operation, inequality 1 is as follows:

[0131]

[0132] θ[n]=E{|C[n]-E(C[n])|}

[0133]

[0134]

[0135] Among them, θ[n], p1[n], λ[n] are parameters used to express their corresponding expressions, and ρ represents a very small value. And combine inequality 1 with the minimum average detection error probability After calculation, we get the following inequality 2:

[0136]

[0137] Among them, θ[n], p1[n], λ[n] are parameters used to express their corresponding expressions, C[n] represents the random variable of the nth time slot, H represents the flight altitude of the drone, β0 represents the channel gain per unit distance, ρ represents a very small value, and P[n] represents the transmission power of the drone and each ground user in the nth time slot. When p2(C[n])=ρ d When

[0138] S4. Propose an optimization problem. The specific expression is as follows:

[0139]

[0140] st

[0141] v[0]=v0,v[N]=v N

[0142] q[0]=q0,q[N]=q N

[0143] ||v[n]||≤v max ,n=1,…,N

[0144] ||v[n]||≥v min,n=1,...,N

[0145] ||a[n]||≤a max ,n=1,…,N

[0146]

[0147] v[n]=v[n-1]+a[n]Δt,n=2,…,N

[0148] in, represents the objective function of optimization, It means maximizing the objective function under the premise that q is the independent variable, q represents the position of the drone, N represents the total number of time slots, and K represents the number of users. represents the variance of the noise uncertainty at the kth user, Δt represents the unit time slot, and h k [n] represents the channel gain at the kth user in the nth time slot, log2 represents the logarithmic function with base 2, q[0] represents the position of the drone at the starting point, q0 represents the initial starting position of the drone, q[N] represents the position of the drone at the end point, q N represents the initialization of the UAV end position, v[0] represents the speed of the UAV at the starting point, v0 represents the initialization of the UAV starting speed, v[N] represents the speed of the UAV at the end point, v N represents the initialization UAV termination speed, log2 represents the logarithmic function with base 2, v[n] represents the flight speed of the UAV at the nth time slot, v[n-1] represents the flight speed of the UAV at the n-1th time slot, v max and v min represents the maximum and minimum values ​​of the UAV's flight speed, a[n] represents the UAV's flight acceleration at the nth time slot, and a max represents the maximum value of the UAV's flight acceleration, st represents the constraint, and P * [n] represents the optimal transmission power of the UAV transmitter, ||v[n]|| and ||a[n]|| represent the Euclidean norm operation of the UAV’s velocity and acceleration in the nth time slot, respectively;

[0149] S5, solving the optimization problem proposed in step S4, specifically comprising the following steps:

[0150] S5.1. Introduce a slack variable constraint and transform the optimization objective function. The specific expression is as follows:

[0151] Slack variable constraints:

[0152] μ k [n]≥||q[n]-w k ||α ,n=1,2,…,N

[0153] The objective function after conversion:

[0154]

[0155] Among them, ||*|| represents the Euclidean norm, α represents the path fading coefficient, q[n] represents the position of the UAV at the nth time slot, H represents the flight altitude of the UAV, β0 represents the channel gain per unit distance, μ k [n] represents a slack variable introduced at the kth user in the nth time slot, w k represents the location of the kth user, represents the variance of the noise uncertainty at the kth user;

[0156] The optimal transmission power P of the UAV transmitter * [n]The expression is as follows:

[0157]

[0158] in, P max Indicates the maximum transmission power of the drone, Represents the smallest of two elements

[0159] S5.2, perform first-order Taylor expansion on the objective function converted in step S5.1 to obtain the following expression:

[0160]

[0161] in, represents the optimized objective function, P * [n] represents the optimal transmission power of the UAV transmitter, H represents the flight altitude of the UAV, μ k [n] represents a slack variable introduced at the kth user in the nth time slot, where is the optimization variable, and μ k [n] is the iteration variable;

[0162] S5.3. Change the constraint of the UAV flight speed in step S5.1. The specific expression is as follows:

[0163]

[0164] Among them, v * [n] represents the flight speed of the drone at the nth time slot, and this variable is an optimization variable, ||*|| 2 represents the square of the Euclidean norm, Indicates v * The transpose of v[n] represents the flight speed of the drone at the nth time slot, v min Indicates the minimum flight speed of the drone;

[0165] S5.4, using the SCA method to solve the transformed optimization problem through iteration and approximation methods;

[0166] The transformed optimization problem is:

[0167]

[0168] st v[0]=v0,v[N]=v N ,

[0169] q[0]=q0,q[N]=q N ,

[0170] ||v[n]||≤v max ,n=1,...,N,

[0171]

[0172] ||a[n]||≤a max ,n=1,…,N,

[0173]

[0174] v[n]=v[n-1]+a[n]Δt,n=2,…,N

[0175] μ k [n]≥||q[n]-w k || α ,n=1,2,…,N

[0176] First, the initial positions of the drone, ground user and listener are q[0],q[F],w k ,w d , initial flight altitude H, total mission duration T, time gap Δt, concealment constraint ρ d , the maximum transmission power P of the UAV transmitter max , channel gain coefficient β0, minimum flight speed v of the drone min , the maximum flight speed of the drone v max , the maximum flight acceleration of the UAV is a max. Secondly, iteratively optimize q[n]: use Matlab's convex optimization modeling system CVX to solve the convex subproblem and get the next iteration point; use the iteration point obtained in the previous step to update the current iteration point, and repeat this process until the maximum number of iterations N is reached or the convergence condition is met. Finally, when the algorithm meets the convergence condition, the optimized value of q[n] is returned, which is q * [n].

[0177] The simulation parameters are shown in Table 1.

[0178] Table 1. System simulation parameters

[0179]

[0180] Figure 2 It shows that the drone starts from the initial point and gradually optimizes the trajectory, flying along the user's position to the final point. During this period, the drone communicates with each user in turn, and the listener eavesdrops on the communication between the drone and the user in real time. Obviously, if the mission time is longer, the drone has the opportunity to get closer to the user to maximize the throughput. As the mission time increases, the drone with the same acceleration has more time to optimize the trajectory, and the drone has the opportunity to get as close as possible to the current communication user to obtain the best communication position and maximize the throughput.

[0181] Figure 2 T=30s is the trajectory of the drone when T=30s in the system described in the embodiment; T=40s is the trajectory of the drone when T=40s in the system described in the embodiment; T=50s is the trajectory of the drone when T=50s in the system described in the embodiment

[0182] Figure 3 It shows that with the increase of concealment performance ρ d The relationship between different UAV flight heights H and total throughput increases. The results show that as the hidden threshold ρ d As the hidden threshold ρ d As the concealment constraint becomes looser, the drone can have greater transmission power and transmit more data to the user. At the same time, different flight altitudes H also have a significant impact on the total throughput between the drone and the user. The total throughput decreases significantly with the increase of H. In addition, as the concealment constraint becomes looser, the gap in total throughput becomes larger. This is because when the drone's flight altitude H climbs, the distance between the drone and the ground user increases. Therefore, the path fading between the drone and the ground user is enhanced, resulting in a decrease in the total throughput.

[0183] Figure 3 Where H = 20m, the total throughput changes with the hidden constraint ρ when H = 20m in the system described in the embodimentd The change curve of H = 50m in the figure is the total throughput with the hidden constraint ρ when H = 50m in the system described in the embodiment d The change curve of H = 80m in the figure is the total throughput with the hidden constraint ρ when H = 80m in the system described in the embodiment d The change curve

[0184] Figure 4 Describes the maximum throughput and maximum transmission power P under different schemes max We can observe that the total throughput increases with the maximum transmission power P max The transmission power of the UAV during flight is not only limited by the covert communication conditions, but also by the maximum transmission power. When the covert communication conditions are met, the UAV can use a larger transmission power P, resulting in a larger transmission power. Therefore, the total throughput increases with P max increases with the increase of . Figure 4 Another illustrative result in is that different mission times T also have a significant impact on the total throughput. A larger mission time T gives the UAV more communication time, thus significantly improving the total throughput between the UAV and the user under a larger mission time T.

[0185] Figure 4 Where T = 30s, the total throughput increases with the maximum transmission power P of the UAV when T = 30s in the system described in the embodiment. max The change curve of the total throughput with the maximum transmission power P of the UAV when T = 40s in the system described in the embodiment; max The change curve of the total throughput with the maximum transmission power P of the UAV when T = 50s in the system described in the embodiment; max The change curve

[0186] The present invention utilizes drones to assist in covert communications and optimizes their trajectories, thereby effectively solving the problems faced by traditional communication methods, such as narrow communication range, inflexible fixed base stations, and signal blocking, and improves the stability and coverage of covert communications. In addition, an approximate convex optimization method is used to successively optimize the variables until the optimal variables are taken.

[0187] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A trajectory optimization method for covert communication assisted by a drone, characterized in that: The following steps are involved: S1. Construct a trajectory optimization system for UAV-assisted covert communication, including a UAV transmitter, multiple legitimate ground user receivers, and a monitor Warden; It also includes two channel links: a covert communication link from the UAV transmitter to each ground user and an eavesdropping link from the UAV transmitter to the listener Warden. The channel gains between the UAV transmitter and the kth ground user and the listener Warden are h d and h k , and the channel gain obeys the complex Gaussian distribution; S2. Perform binary hypothesis test calculation on the listener Warden to obtain the false alarm probability P r1 and the probability of missed detection P r2 , and calculate the total detection error rate ξ of the listener Warden, and finally get the optimal average minimum error detection probability ξ of the listener Warden * ; S3. Analyze the trajectory of the drone and combine it with the optimal average minimum error detection probability ξ * Optimize the flight trajectory of the drone and ultimately obtain the optimal transmission power of the drone; S4. Propose an optimization problem. The specific expression is as follows: in, represents the objective function of optimization, It means maximizing the objective function under the premise that q is the independent variable, q represents the position of the drone, N represents the total number of time slots, and K represents the number of users. represents the variance of the noise uncertainty at the kth user, Δt represents the unit time slot, and h k [n] represents the channel gain at the kth user at the nth time slot, log2 represents the logarithmic function with base 2, v[n] represents the flight speed of the drone at the nth time slot, v[n-1] represents the flight speed of the drone at the n-1th time slot, and v max and v min represents the maximum and minimum values ​​of the UAV's flight speed, a[n] represents the UAV's flight acceleration at the nth time slot, and a max represents the maximum value of the UAV's flight acceleration, st represents the constraint, and P * [n] represents the optimal transmission power of the UAV transmitter, ||*||; S5, solving the optimization problem proposed in step S4, specifically comprising the following steps: S5.

1. Introduce a slack variable constraint and transform the optimization objective function. The specific expression is as follows: Slack variable constraints: μ k [n]≥||q[n]-w k || α ,n=1,2,…,N The objective function after conversion: Among them, ||*|| represents the Euclidean norm, α represents the path fading coefficient, q[n] represents the position of the UAV at the nth time slot, H represents the flight altitude of the UAV, and μ k [n] represents a slack variable introduced at the kth user in the nth time slot, w k represents the position of the kth user, β0 represents the channel gain per unit distance, represents the variance of the noise uncertainty at the kth user; S5.2, perform first-order Taylor expansion on the objective function converted in step S5.1 to obtain the following expression: in, represents the optimized objective function, P * [n] represents the optimal transmission power of the UAV transmitter, H represents the flight altitude of the UAV, μ k [n] represents a slack variable introduced at the kth user in the nth time slot, where is the optimization variable, and μ k [n] is the iteration variable; S5.

3. Change the constraint of the UAV flight speed in step S5.

1. The specific expression is as follows: Among them, v * [n] represents the flight speed of the drone at the nth time slot, ||*|| 2 represents the square of the modulus of *, Indicates v * The transpose of v[n] represents the flight speed of the drone at the nth time slot, v min Indicates the minimum flight speed of the drone; S5.

4. Use the SCA method to solve the transformed optimization problem through iteration and approximation methods.

2. The trajectory optimization method for UAV-assisted covert communication according to claim 1 is characterized in that: The step S2 comprises the following steps: S2.

1. Obtain the drone that Warden eavesdropped on in the nth time slot to send a covert signal y to the ground user d [n], its specific expression is as follows: Among them, x[n] represents the signal sent by the drone in the nth time slot, H0 represents the null hypothesis, H1 represents the alternative hypothesis, and n d [n] represents the additive Gaussian white noise interference signal received by the listener in the nth time slot, and its mean is 0 and the variance is The normal distribution of y d [n] indicates that the eavesdropper Warden eavesdrops on the hidden signal in the nth time slot, P[n] indicates the transmission power of no one at this time, h d represents the channel fading coefficient between the UAV transmitter and the listener, Follow uniform distribution; S2.2, the power expression of the signal received by the listener Warden is as follows: Among them, P d [n] represents the power of the signal received by the listener Warden, h d [n] represents the channel fading coefficient between the transmitter of the drone and the listener Warden in the nth time slot, H0 represents the null hypothesis, H1 represents the alternative hypothesis, and P[n] represents the transmission power of the drone at this time; S2.

3. The listener Warden uses a binary hypothesis test based on the received signal power to determine whether the sender is communicating covertly with the ground user. D0 and D1 represent the listener's decision to determine whether the sender is communicating covertly with the ground user. D0 means that the listener Warden believes that the sender is not sending covert information, and D1 means that the listener Warden believes that the sender is communicating covertly with the ground user. The expression of the total detection error probability ξ[n] of the listener Warden in the nth time slot is as follows: ξ[n]=P r1 [n]+P r2 [n] Where ξ[n] represents the total detection error probability of the nth time slot listener Warden, P r1 [n] represents the false alarm probability of the nth time slot, P r2 [n] represents the missed detection probability of the nth time slot, and the expressions are as follows: Among them, P r0 [n] represents the false alarm probability of the nth time slot, Express satisfaction The probability of Express satisfaction The probability, P de [n] represents the error detection threshold of the listener Warden at the nth time slot, h d [n] represents the channel fading coefficient between the transmitting end of the drone and the listener Warden in the nth time slot, and P[n] represents the transmission power between the drone and each ground user in the nth time slot; S2.

4. Calculate the optimal average minimum error detection probability ξ of the listener * [n], the specific expression is as follows: Among them, ξ * [n] represents the minimum total detection error probability of the listener in the nth time slot, h d [n] represents the channel fading coefficient between the drone transmitter and the listener Warden in the nth time slot, and P[n] represents the transmission power between the drone and each ground user in the nth time slot.

3. The trajectory optimization method for UAV-assisted covert communication according to claim 2 is characterized in that: The false alarm probability P in step S2.3 r1 The expression is as follows: Among them, P r1 represents the false alarm probability of the monitor Warden, P de [n] represents the error detection threshold of the listener Warden at the nth time slot, σ 2 is the noise uncertainty at Warden The distribution parameters obeyed, The obedience parameter is [σ 2 -ρ,σ 2 +ρ] uniform distribution, ρ represents a minimum value; Missed detection probability P r2 The expression is as follows: Where P[n] represents the transmission power between the drone and each ground user in the nth time slot, Q[n] = h d [n]P[n],P de [n] represents the error detection threshold of the listener Warden at the nth time slot, σ 2 is the noise uncertainty at Warden The distribution parameters obeyed, The obedience parameter is [σ 2 -ρ,σ 2 +ρ] uniform distribution, ρ represents a minimum value; The specific expression of the total detection error probability ξ[n] of the nth time slot listener Warden is as follows: Among them, P de [n] represents the error detection threshold of the listener Warden at the nth time slot, σ 2 is the noise uncertainty at Warden The distribution parameters obeyed, The obedience parameter is [σ 2 -ρ,σ 2 +ρ], ρ represents a minimum value, P[n] represents the transmission power of the UAV and the ground users in the nth time slot, h d [n] represents the channel fading coefficient between the UAV transmitter and the listener Warden in the nth time slot.

4. The trajectory optimization method for UAV-assisted covert communication according to claim 3 is characterized in that: The specific process of step S3 is as follows: S3.

1. Introduce a random variable C[n] that reflects the channel uncertainty between the drone and the listener in the actual scenario; S3.2, based on the introduced random variable C[n] and combined with the optimal average minimum error detection probability ξ * [n], get the minimum average detection error probability Minimum average detection error probability The expression is as follows: Among them, E C[n] (*) represents the expectation of C[n], ξ * (C[n]) represents the minimum detection error probability with C[n] as the independent variable, H represents the flight altitude of the drone, represents the value used to simulate the uncertainty of the channel between the drone and the listener, and β0 represents the channel gain per unit distance; S3.3, the random variable C[n] is subject to Gaussian distribution, that is, C[n]~N(λ[n]+2,4λ[n]+4), and the following inequality is combined with the minimum average detection error probability Performing the operation, inequality 1 is as follows: θ[n]=E{|C[n]-E(C[n])|} Among them, θ[n], p1[n], λ[n] are parameters used to express their corresponding expressions. represents the uncertainty value used to simulate the channel between the drone and the listener, ρ represents a minimum value, And combine inequality 1 with the minimum average detection error probability After calculation, we get the following inequality 2: Among them, θ[n], p1[n], λ[n] are parameters used to express their corresponding expressions, C[n] represents the random variable of the nth time slot, H represents the flight altitude of the drone, β0 represents the channel gain per unit distance, ρ represents a minimum value, and P[n] represents the transmission power of the drone and each ground user in the nth time slot. When p2(C[n])=ρ d When , the transmitting power of the UAV transmitter reaches the maximum value.

5. The trajectory optimization method for UAV-assisted covert communication according to claim 4 is characterized in that: The specific expression in step S3.1 is as follows: Among them, ||*|| represents the Euclidean norm, α represents the path fading coefficient, q[n] represents the position of the UAV at the nth time slot, and C[n] represents the random variable of the nth time slot. represents the uncertainty value used to simulate the channel between the drone and the listener, w d Indicates the location of the Warden.

6. The trajectory optimization method for UAV-assisted covert communication according to claim 4 is characterized in that: The optimal transmission power P of the UAV transmitter * [n]The expression is as follows: in, P max Indicates the maximum transmission power of the drone, Represents the smallest of two elements.

Citation Information

Cited By

  • Track design method and system in unmanned aerial vehicle hidden and inductive integrated network

    CN120915373A