Minimum safe rate maximization method for UAV-assisted NOMA backscatter communication system

Abstract

The present invention claims a method for maximizing the minimum safe rate of a drone-assisted non-orthogonal multiple access technology NOMA backscatter communication system. The method takes into account the fairness between backscatterers BD and the difficulty of eavesdroppers to obtain instantaneous channel state information CSI, and satisfies the transmission power of the drone, the decoding order of serial interference cancellation SIC, and the reflection coefficient of BD, energy collection EH and the safety interruption probability constraints. The method realizes the maximization of the minimum safe rate of the drone-assisted NOMA backscatter communication system. The present invention first processes the safety interruption probability constraint by referencing auxiliary variables, and then obtains the optimal solution by using methods such as block coordinate descent, variable substitution, penalty function, scaling and continuous convex approximation SCA. The present invention takes into account the fairness between BDs and the actual eavesdropping scenario at the same time, and has good practicality and feasibility.

Description

Technical Field

[0001] The present invention belongs to the technical field of resource allocation of unmanned aerial vehicle-assisted NOMA backscatter communication system, and specifically relates to a method for maximizing the minimum safe rate of a unmanned aerial vehicle-assisted NOMA backscatter communication system. Background Art

[0002] From the perspective of the 6G vision of "Internet of Everything, Intelligent Twins", a large number of devices will be connected to the communication network, and communication between things will become a research hotspot. In the traditional Internet of Things, the energy of wireless devices is limited by their battery capacity, and regularly replacing the battery of sensors or charging them will increase costs. Therefore, backscatter communication technology with low cost, low power consumption and low complexity has broad development prospects in the Internet of Things. However, due to two-way attenuation, the signal received by the backscatter receiver (BR) is very weak, and the available part of the environmental energy is limited, so the backscatter communication range is limited. In recent years, drones have become the focus of scientific researchers due to their high maneuverability, flexible deployment and good line of sight with ground users. Introducing drones into backscatter communication systems can increase the communication distance. In the resource allocation research of drone-assisted backscatter communication systems, the communication protocol between BD and BR is mostly time division multiple access (TDMA). However, with the rapid development of 6G and the Internet of Things, TDMA has too low spectrum efficiency in the scenario of large-scale Internet of Everything. Therefore, it is necessary to introduce NOMA technology that can accommodate more users and has higher spectrum efficiency and energy efficiency. Similar to other wireless systems, backscatter communication systems are vulnerable to potential eavesdropping attacks, so physical layer security is also a key issue to consider in backscatter communication systems.

[0003] At present, in the resource allocation method of UAV-assisted backscatter communication system, HU J et al. published an article entitled "Joint trajectory and scheduling design for UAV aided secure backscatter communications" in "IEEE Wireless Communications Letters, 2020, 9(12): 2168-2172.", which studied the problem of maximizing the fair and secure rate of the uplink with the trajectory of the UAV, the scheduling of BD and its reflection coefficient as optimization variables; BAI L et al. published an article entitled "UAV-enabled secure multiuser backscatter communications with planar array" in "IEEE Journal on Selected Areas in Communications, 2022, 40(10): 2946-2961.", which studied the problem of maximizing the secure rate of a multi-user backscatter communication system assisted by a UAV using analog beamforming and random continuous wave technology.

[0004] From the above research results, it can be seen that most scholars have studied the resource allocation based on maximizing the security rate in the drone-assisted NOMA backscatter communication system. The communication protocol between BD and BR is mostly TDMA, and the NOMA protocol is used less frequently. In addition, only the communication environment in which the eavesdropper's channel is an ideal CSI is considered. In the actual eavesdropping scenario, the eavesdropper usually remains silent during eavesdropping, and it is difficult to obtain the instantaneous CSI of the eavesdropping channel. Therefore, the present invention considers the fairness between BDs and the difficulty in obtaining the instantaneous CSI of the eavesdropper, and studies the minimum security rate maximization method of the drone-assisted NOMA backscatter communication system. Summary of the invention

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

[0006] A method for maximizing the minimum safe rate of a drone-assisted NOMA backscatter communication system comprises the following steps:

[0007] 101. Establish a minimum safety rate maximization model for the drone-assisted NOMA backscatter communication system, use the safety interruption probability as a safety metric, and use auxiliary variables to handle the safety interruption probability constraint;

[0008] 102. Initialize the relevant parameters of the target problem, including the number of inner iterations, the number of outer iterations, the maximum number of outer iterations, the penalty parameter, the UAV transmission power p, the reflection coefficient R of the BD, the variable T, the variable θ, the UAV position q, the SIC decoding order A, the tolerance factor and the dual variable;

[0009] 103. According to the given UAV position, SIC decoding order, BD reflection coefficient and variable T, the UAV transmission power is solved and updated;

[0010] 104. According to the given UAV transmission power, UAV position, SIC decoding order and variable T, the reflection coefficient of BD is solved and updated;

[0011] 105. According to the given UAV transmission power, BD reflection coefficient, UAV position and SIC decoding order, solve the variable T and update it;

[0012] 106. According to the given UAV transmission power and BD reflection coefficient, the UAV position and SIC decoding order are solved and updated;

[0013] 107. Minimum safe rate update convergence judgment: If the absolute value of the difference between the two safe rates is not greater than the tolerance factor Then execute step 108; if the absolute value of the difference between the two safety speeds is greater than the tolerance factor The current safety rate is saved and the process goes to step 103 until the safety rate meets the condition.

[0014] 108. Penalty parameter convergence judgment: use the q and θ output from the inner loop to calculate If the penalty function Not greater than tolerance factor And ||y|| ∞ Not greater than tolerance factor Then update the outer iteration number and dual variables; otherwise, update the penalty parameter and jump to step 103;

[0015] 109. Compare the outer layer iteration number with the outer layer maximum iteration number. If it is greater than the outer layer maximum iteration number, the algorithm ends; otherwise, jump to step 103 until the condition is met.

[0016] The advantages and beneficial effects of the present invention are as follows:

[0017] The present invention aims to maximize the minimum safe rate between BDs under the constraints of the transmission power of the drone, the SIC decoding order, the reflection coefficient of the BD, EH and the probability of safe interruption. Since the proposed optimization problem is non-convex and difficult to solve directly, the present invention proposes an inner and outer two-layer iterative algorithm based on methods such as block coordinate descent and SCA for solving it. Its innovation lies in that in step 101, while considering the fairness between BDs and the difficulty of obtaining the instantaneous CSI of the eavesdropper, a minimum safe rate maximization model for the drone-assisted NOMA backscatter communication system is established, and auxiliary variables are used to process the probability of safe interruption constraints. In step 106, since the optimization subproblem is non-convex and it is difficult to solve it directly, in order to facilitate calculation, penalty functions and SCA methods are used to convert non-convex optimization subproblems into convex optimization subproblems. Compared with other schemes, the method provided by the present invention has the advantages of low computational complexity, and is more practical and feasible. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 The present invention provides a system model of a preferred embodiment of a drone-assisted NOMA backscatter communication system;

[0019] Figure 2 It is an iterative convergence diagram of the minimum safety rate of the system of the present invention;

[0020] Figure 3 The minimum safe rate of the system under different maximum transmission powers of UAVs for the present invention and the comparison method;

[0021] Figure 4 The minimum safe rate of the system at different UAV altitudes for the present invention and the comparison method;

[0022] Figure 5 The minimum safe rate of the system under different numbers of BDs for the present invention and the comparative method;

[0023] Figure 6 A schematic flow chart of a method for maximizing the minimum safety rate of a drone-assisted NOMA backscatter communication system is provided for a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0024] The following will describe the technical solutions in the embodiments of the present invention in detail in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention.

[0025] The technical solution of the present invention to solve the above technical problems is:

[0026] Figure 6A method for maximizing the minimum safe rate of a drone-assisted NOMA backscatter communication system is disclosed, which includes the following steps:

[0027] Step 1: Considering the fairness between BDs and the difficulty of eavesdroppers to obtain the instantaneous CSI, a minimum security rate maximization model for the UAV-assisted NOMA backscatter communication system is established, and the security interruption probability is used as the security metric, and the security interruption probability constraint is processed using auxiliary variables;

[0028] Step 2: Initialize the relevant parameters of the target problem, the number of inner iterations, the number of outer iterations, the maximum number of outer iterations, the penalty parameter, the UAV transmission power p, the reflection coefficient R of the BD, the variable T, the variable θ, the UAV position q, the SIC decoding order A, the tolerance factor, and the dual variable;

[0029] Step 3: Based on the given UAV position, SIC decoding order, BD reflection coefficient, and variable T, solve the UAV transmission power and update it;

[0030] Step 4: Based on the given UAV transmission power, UAV position, SIC decoding order, and variable T, solve the reflection coefficient of BD and update it;

[0031] Step 5: According to the given UAV transmission power, BD reflection coefficient, UAV position, SIC decoding order, solve the variable T and update it;

[0032] Step 6: Based on the given UAV transmission power and BD reflection coefficient, the UAV position and SIC decoding order are solved and updated;

[0033] Step 7: Minimum safe rate update convergence judgment: If the absolute value of the difference between the two safe rates is not greater than the tolerance factor Then execute step 108; if the absolute value of the difference between the two safety speeds is greater than the tolerance factor The current safety rate is saved, and the process jumps to step 103 until the safety rate meets the condition.

[0034] Step 8: Penalty parameter convergence judgment: use the q and θ output from the inner loop to calculate If the penalty function Not greater than tolerance factor And ||y|| ∞ Not greater than tolerance factor Then update the outer iteration number and dual variable; otherwise, update the penalty parameter and jump to step 103.

[0035] Step 9: Compare the outer layer iteration number with the outer layer maximum iteration number. If it is greater than the outer layer maximum iteration number, the algorithm ends; otherwise, jump to step 103 until the condition is met.

[0036] Furthermore, in the first step, considering the fairness between BDs and the difficulty of eavesdroppers to obtain instantaneous CSI, the minimum security rate maximization objective optimization problem of the drone-assisted NOMA backscatter communication system is established as follows:

[0037]

[0038] In the formula, p u represents the transmission power of the drone, q represents the position of the drone, R represents the reflection coefficient of BD, A represents the SIC decoding order, R s,n represents the safety rate of the nth BD, r n ∈[0,1] represents the reflection coefficient of the nth BD, η n ∈[0,1] represents the energy efficiency conversion coefficient of the nth BD, h n is the channel power gain from UAV to BD, R n represents the rate from the nth BD to the UAV, R nE represents the rate from the nth BD to the eavesdropper, α nm ∈{0,1} represents the SIC decoding order, α nn = 0 means that the UAV should not regard the signal of the nth BD as interference when decoding the signal of the nth BD, H represents the flight altitude of the UAV, and w n represents the position of the nth BD, N represents the number of BDs; C1a represents the transmission power constraint of the UAV, P max represents the maximum transmission power of the UAV; C1b represents the reflection coefficient r n Constraint; C1c is the EH constraint, which means that the energy collected by BD is not less than the energy it consumes, P c is the power consumed by BD to maintain its own circuit operation; C1d is the SOP constraint of BD, p max,n represents the probability size; C1e-C1h represents the SIC decoding constraints, where C1e represents that for two different BDs, α nm is 0 or 1; C1g means that for two different BDs, if one is considered to be a BD with a weaker channel gain, the other must be a BD with a stronger channel gain; C1e, C1g and C1h ensure that when the channel gain of the mth BD is weaker than that of the nth BD, there is α nm =1, otherwise α nm =0.

[0039] Specifically, the rates of the nth BD to the drone and the eavesdropper can be expressed as:

[0040]

[0041]

[0042] Where h n is the channel power gain from UAV to BD, g n is the channel power gain from the eavesdropper to BD, σ 2 is the Gaussian noise at the BD or eavesdropper.

[0043] Introducing slack variables Transform constraint C1d into:

[0044]

[0045] And t n satisfy:

[0046]

[0047] make d mE and d nE They represent the distances from the mth BD and the nth BD to the eavesdropper, β0 represents the channel power gain at the reference distance d0 = 1m, and

[0048]

[0049] ζ n and m represents an exponentially distributed random variable with unit mean.

[0050] So the target optimization problem p1 can be converted into:

[0051]

[0052] Furthermore, in the second step, the relevant parameters of the target problem are initialized, including the number of inner iterations, the number of outer iterations, the maximum number of outer iterations, the penalty parameter, the UAV transmission power p, the reflection coefficient R of the BD, the variable T, the variable θ, the UAV position q, the SIC decoding order A, the penalty factor, and the dual variable.

[0053] Furthermore, in the third step, given the reflection coefficient of BD, variable T, drone position, SIC decoding order, the drone transmission power is optimized as follows:

[0054] Introducing auxiliary variables s, the target optimization problem p2 is transformed into:

[0055]

[0056] Among them, s is the introduced slack variable. C3a represents the transmission power constraint of the UAV, P max represents the maximum transmission power of the UAV. C3b is the EH constraint, which means that the energy collected by BD is not less than the energy consumed by it. c The power consumed by BD to maintain its own circuit operation. Constraint C3c is The constraint is obtained by taking the logarithm of both sides of the inequality. Constraint C3d is a constraint on the newly introduced variable s. The SCA method is used to calculate the initial feasible point. The constraints are obtained after the first-order Taylor expansion.

[0057] Furthermore, in the fourth step, given the drone transmission power, variable T, drone position, SIC decoding order, the BD reflection coefficient is optimized, specifically:

[0058] Introducing auxiliary variables u, the target optimization problem p2 is transformed into:

[0059]

[0060] in, Constraint C4c is obtained by taking the logarithm of both sides of the inequality of constraint C2d and using the SCA method to find the initial feasible point. and The constraint C4d is obtained by first rewriting the constraint of the newly introduced variable u as follows: Again Using SCA method to find the initial feasible point It is obtained by performing a first-order Taylor expansion at .

[0061] Furthermore, in the fifth step, given the UAV transmission power, BD reflection coefficient, UAV position, SIC decoding order, the optimized variable T is specifically:

[0062] Introducing auxiliary variables l, the target optimization problem p2 is transformed into:

[0063]

[0064] Among them, constraint C5a is obtained by taking the logarithm of both sides of the C2d inequality; constraint C5b is a newly introduced constraint The left side uses the SCA method at the initial feasible point It is obtained by performing a first-order Taylor expansion at .

[0065] Furthermore, in the sixth step, given the drone transmission power, the reflection coefficient of BD, the variable T, the optimized drone position, and the SIC decoding order are specifically:

[0066] Introducing auxiliary variables The target optimization problem p2 is transformed into:

[0067]

[0068] In order to facilitate calculation, the equality constraint is introduced. Using the penalty dual decomposition algorithm, p6 can be transformed into the following optimization problem:

[0069]

[0070] in, τ is the penalty parameter; is the dual variable of the equality constraint. By making the discrete binary variable A continuous, constraint C7c is equivalent to: Introduce a penalty function into the objective function to relax the constraint: And get the new constraint: Represents the introduced penalty function.

[0071] And the transformed optimization problem can be rewritten as:

[0072]

[0073] in, It is an extended constraint The set of slack variables in the feasible region; λ>0 represents the penalty parameter. For C8h, let And introduce slack variables and C8h can be rewritten as the following expression:

[0074]

[0075]

[0076]

[0077] For C8h3, the slack variable is introduced and C8h3 can be rewritten as follows:

[0078]

[0079]

[0080]

[0081] By making the non-convex constraints convex, we can get the optimization problem:

[0082]

[0083] Among them, C9f is to use the SCA method to constrain C8d at the initial feasible point C9g is obtained by rewriting constraint C8g as Then, the SCA method is used on the left side to find the initial feasible point and The constraint C9h is obtained by taking the logarithm of both sides of constraint C8b and then calculating -ln(r n ||w m -e|| 2 θ m ) Using the SCA method at the initial feasible point The constraint C9i is obtained by using the SCA method on the left side of the constraint C8h1 at the initial feasible point and The constraint C9j is obtained by expanding the constraint C8h2 using the SCA method at the initial feasible point The constraint C9k is obtained by using the SCA method at the initial feasible point of constraint C8h4. Constraint C9l is obtained by using the SCA method at the initial feasible point of constraint C8h5. Constraint C9m is obtained by using the SCA method at the initial feasible point of constraint C8h6. It is obtained by the first-order Taylor expansion.

[0084] Furthermore, in the seventh step, compare With tolerance factor The size of is the maximum minimum safe rate for the n+1th iteration, is the maximum minimum safe rate of the nth iteration; if No more than The safety rate converges, and the eighth step is executed; if the absolute value of the difference between the two safety rates is greater than the tolerance factor The current safety rate is saved and the process jumps to the third step until the safety rate meets the conditions.

[0085] Furthermore, in the eighth step, the q and θ output from the inner loop are used to calculate If the penalty function Not greater than tolerance factor And ||y|| ∞ Not greater than tolerance factor Then update the outer iteration number and dual variables; otherwise, update the penalty parameter and jump to the third step.

[0086] Furthermore, in the ninth step, the outer layer iteration number is compared with the outer layer maximum iteration number. If it is greater than the outer layer maximum iteration number, the algorithm ends; otherwise, jump to the third step until the conditions are met.

[0087] The invention considers fairness between BDs and the difficulty of eavesdroppers to obtain instantaneous CSI, and satisfies the transmission power of the drone, the SIC decoding order, the reflection coefficient of the BD, the EH, and the safety interruption probability constraints. By referencing auxiliary variables to process the safety interruption probability constraints, the non-convex optimization problem is equivalently converted into a convex optimization problem based on variable substitution, penalty function, scaling, and continuous convex approximation, and then an alternating iterative algorithm is used to maximize the minimum safety rate between BDs. The invention has the advantages of low computational complexity, while considering fairness between BDs and actual eavesdropping scenarios, and has better practicality and feasibility.

[0088] This embodiment is a method for maximizing the minimum safe rate of a drone-assisted NOMA backscatter communication system. In the drone-assisted NOMA backscatter communication system, BDs are randomly distributed in a 40m×40m square area, the maximum number of BDs is 8, the drone flight altitude is 10m-15m, the maximum transmission power of the drone is 0.2W-1.0W, the reference channel power gain β0=-30dB, the EE conversion coefficient η=0.6, and the noise power δ 2 =-100dBm, BD circuit power p c =0.25uW, the channel from the drone to the BD is modeled as a line-of-sight channel, and the channel from the eavesdropper to the BD is modeled as a Rayleigh fading channel.

[0089] In this example, Figure 1 The system model of the drone-assisted NOMA backscatter communication system of the preferred embodiment of the present invention includes a full-duplex drone, N BDs and an eavesdropper, wherein the eavesdropper can steal the signal reflected by the BD to the drone and the instantaneous CSI of the eavesdropper is difficult to obtain; Figure 2 It is an iterative convergence diagram of the minimum safety rate of the system of the present invention; Figure 3 The minimum safe rate of the system for the present invention and the comparison method (geometric center, OMA, random position) under different maximum transmission powers of the UAV; Figure 4 The minimum safe rate of the system at different UAV altitudes for the present invention and the comparison method; Figure 5 The minimum safe rate of the system under different BD numbers for the present invention and the comparison method. Figure 2 It can be seen that the method proposed in the present invention increases with the increase of the number of iterations and tends to converge after 8 iterations. Figure 3It can be seen that the proposed method is superior to the comparison method, and the minimum safe rate of the proposed method increases with the increase of the maximum transmit power of the UAV. Under high transmit power conditions, the minimum safe rate changes little. Figure 4 It can be seen that the minimum safe rate of the system decreases with the increase of the UAV's flight altitude, and the minimum safe rate of the proposed method is higher than that of the comparison method. Figure 5 It can be seen that with the increase of the number of BDs, the minimum safe rate of the system is decreasing, and the minimum safe rate of the proposed method is higher than that of the comparison method.

[0090] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions.

[0091] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0092] The above embodiments should be understood to be only used to illustrate the present invention and not to limit the protection scope of the present invention. After reading the contents of the present invention, technicians can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. A method for maximizing the minimum safe rate of a drone-assisted NOMA backscatter communication system, characterized in that: The following steps are involved:

101. Considering the fairness between BDs and the difficulty of eavesdroppers to obtain instantaneous CSI, a minimum security rate maximization model for drone-assisted NOMA backscatter communication system is established, and the security interruption probability is used as a security metric, and auxiliary variables are used to handle the security interruption probability constraint; 102. Initialize the relevant parameters of the target problem, including the number of inner iterations, the number of outer iterations, the maximum number of outer iterations, the penalty parameter, the UAV transmission power p, the reflection coefficient R of the BD, the variable T, the variable θ, the UAV position q, the SIC decoding order A, the tolerance factor and the dual variable; 103. According to the given UAV position, SIC decoding order, BD reflection coefficient and variable T, the UAV transmission power is solved and updated; 104. According to the given UAV transmission power, UAV position, SIC decoding order and variable T, the reflection coefficient of BD is solved and updated; 105. According to the given UAV transmission power, BD reflection coefficient, UAV position and SIC decoding order, solve the variable T and update it; 106. According to the given UAV transmission power and BD reflection coefficient, the UAV position and SIC decoding order are solved and updated; 107. Minimum safe rate update convergence judgment: If the absolute value of the difference between the two safe rates is not greater than the tolerance factor Then execute step 108; if the absolute value of the difference between the two safety speeds is greater than the tolerance factor The current safety rate is saved and the process goes to step 103 until the safety rate meets the condition.

108. Penalty parameter convergence judgment: use the q and θ output of the inner loop to calculate If the penalty function Not greater than tolerance factor And ||y|| ∞ Not greater than tolerance factor Then update the outer iteration number and dual variables; otherwise, update the penalty parameter and jump to step 103; 109. Compare the outer layer iteration number with the outer layer maximum iteration number. If it is greater than the outer layer maximum iteration number, the algorithm ends; otherwise, jump to step 103 until the condition is met; The step 101 establishes the minimum safety rate maximization objective optimization problem of the drone-assisted NOMA backscatter communication system as follows: In the formula, p u represents the transmission power of the drone, q represents the position of the drone, R represents the reflection coefficient of BD, A represents the SIC decoding order, R s,n represents the safety rate of the nth BD, r n ∈[0,1] represents the reflection coefficient of the nth BD, η n ∈[0,1] represents the energy efficiency conversion coefficient of the nth BD, h n is the channel power gain from UAV to BD, R n represents the rate from the nth BD to the UAV, R nE represents the rate from the nth BD to the eavesdropper, α nm ∈{0,1} represents the SIC decoding order, α nn = 0 means that the UAV should not regard the signal of the nth BD as interference when decoding the signal of the nth BD, H represents the flight altitude of the UAV, and w n represents the position of the nth BD, N represents the total number of BDs; C1a represents the transmission power constraint of the UAV, P max represents the maximum transmission power of the UAV; C1b represents the reflection coefficient r n Constraint; C1c is the EH constraint, which means that the energy collected by BD is not less than the energy it consumes, P c is the power consumed by BD to maintain its own circuit operation; C1d is the safety interruption probability SOP constraint of BD, p max,n represents the probability size; C1e-C1h represents the SIC decoding constraints, where C1e represents that for two different BDs, α nm is 0 or 1; C1g means that for two different BDs, if one is considered to be a BD with a weaker channel gain, the other must be a BD with a stronger channel gain; C1e, C1g and C1h ensure that when the channel gain of the mth BD is weaker than that of the nth BD, there is α nm =1, otherwise α nm =0; The rates of the nth BD to the drone and the eavesdropper can be expressed as: Where h n is the channel power gain from UAV to BD, g n is the channel power gain from the eavesdropper to BD, σ 2 is the Gaussian noise at the BD or eavesdropper; Introducing slack variables The safety interruption probability constraint C1d is transformed into: And t n satisfy: make Among them, d mE and d nE denote the distances from the mth BD and the nth BD to the eavesdropper, β0 denotes the channel power gain at the reference distance d0 = 1m, then ζ n and m represents a random variable with an exponential distribution and unit mean; So the target optimization problem p1 can be converted into:

2. The method for maximizing the minimum safe rate of a drone-assisted NOMA backscatter communication system according to claim 1 is characterized in that: The reflection coefficient of the BD, the variable T, the position of the UAV, the SIC decoding order, and the optimized UAV transmission power given in step 103 are specifically: Introducing auxiliary variables s, the target optimization problem p2 is transformed into: Among them, s is the introduced slack variable, C3a represents the transmission power constraint of the UAV, C3b is the EH constraint, which means that the energy collected by BD is not less than the energy consumed by it, and P c is the power consumed by BD to maintain its own circuit operation. Constraint C3c is the constraint obtained by taking the logarithm of both sides of the inequality C2d. Constraint C3d is the constraint on the newly introduced variable s. Use the SCA method to calculate the initial feasible point p u It is obtained after the first-order Taylor expansion.

3. The method for maximizing the minimum safe rate of a drone-assisted NOMA backscatter communication system according to claim 2 is characterized in that: The step 104, given the UAV transmission power, variable T, UAV position, and SIC decoding order, optimizes the BD reflection coefficient, specifically: Introducing auxiliary variables u, the target optimization problem p2 is transformed into: in, Constraint C4c is obtained by taking the logarithm of both sides of the inequality of constraint C2d and using the SCA method to find the initial feasible point. and The constraint C4d is obtained by first rewriting the constraint of the newly introduced variable u as follows: Again It is obtained by performing a first-order Taylor expansion at the initial feasible point u using the SCA method.

4. The method for maximizing the minimum safe rate of a drone-assisted NOMA backscatter communication system according to claim 3 is characterized in that: In step 105, given the UAV transmission power, the BD reflection coefficient, the UAV position, and the SIC decoding order, the variable T is optimized, specifically: Introducing auxiliary variables l, the target optimization problem p2 is transformed into: Among them, constraint C5a is obtained by taking the logarithm of both sides of the C2d inequality; constraint C5b is a newly introduced constraint The left side uses the SCA method at the initial feasible point It is obtained by performing a first-order Taylor expansion at .

5. The method for maximizing the minimum safe rate of a drone-assisted NOMA backscatter communication system according to claim 4 is characterized in that: The step 106, given the UAV transmission power, the reflection coefficient of BD, and the variable T, optimizes the UAV position and the SIC decoding order, specifically: Introducing auxiliary variables The target optimization problem p2 is transformed into: Introducing equality constraints Use the penalty dual decomposition algorithm to transform p6 into the following optimization problem: in, τ is the penalty parameter; is the dual variable of the equality constraint; the discrete binary variable A is made continuous, and the constraint C7c is equivalent to: Introduce a penalty function into the objective function to relax the constraint: And get the new constraint: Represents the introduced penalty function; And the transformed optimization problem is rewritten as: in, It is an extended constraint The set of slack variables in the feasible region; λ>0 represents the penalty parameter; for C8h, let And introduce slack variables and C8h can be rewritten as the following expression: For C8h3, the slack variable is introduced and Rewrite C8h3 as follows: By making the non-convex constraints convex, we get the following optimization problem: Among them, C9f is to use the SCA method to constrain C8d at the initial feasible point C9g is obtained by rewriting constraint C8g as Then, the SCA method is used on the left side to find the initial feasible point and The constraint C9h is obtained by taking the logarithm of both sides of constraint C8b and then calculating -ln(r n ||w m -e|| 2 θ m ) Using the SCA method at the initial feasible point The constraint C9i is obtained by using the SCA method on the left side of the constraint C8h1 at the initial feasible point and The constraint C9j is obtained by expanding the constraint C8h2 using the SCA method at the initial feasible point The constraint C9k is obtained by using the SCA method at the initial feasible point of constraint C8h4. Constraint C9l is obtained by using the SCA method at the initial feasible point of constraint C8h5. Constraint C9m is obtained by using the SCA method at the initial feasible point of constraint C8h6. It is obtained by the first-order Taylor expansion.

6. The method for maximizing the minimum safe rate of a drone-assisted NOMA backscatter communication system according to claim 5 is characterized in that: The step 107, comparing With tolerance factor The size of is the maximum minimum safe rate for the n+1th iteration, is the maximum minimum safe rate of the nth iteration; if No more than The safety rate converges, and step 108 is executed; if the absolute value of the difference between the two safety rates is greater than the tolerance factor The current safety rate is saved, and the process jumps to step 103 until the safety rate meets the condition.

Images