Multi-unmanned aerial vehicle relay auxiliary communication rate optimization method based on MC-NOMA

Abstract

The invention belongs to the technical field of wireless communication, and particularly relates to a multi-unmanned aerial vehicle relay auxiliary communication rate optimization method based on MC-NOMA, and the method comprises the following steps: S1, building an air-to-ground channel model and a power consumption model according to the characteristics of a multi-unmanned aerial vehicle auxiliary wireless communication scene; s2, performing problem modeling according to user access grouping constraints, physical constraints and unmanned aerial vehicle performance constraints; and S3, decomposing an original multi-objective optimization problem with priority into a single-objective optimization problem by utilizing a sequential optimization principle. According to the method, through a method of combining sequential optimization, block coordinate descent and successive convex approximation, the problems of low user communication rate, fixed user access mode, poor adaptive channel capability, high energy consumption and the like in multi-unmanned aerial vehicle assisted wireless communication are solved.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a multi-UAV relay-assisted communication rate optimization method based on MC-NOMA. Background Art

[0002] With the in-depth development of Multi-UAV Relay-Assisted Communication technology in scientific research and practical applications, its flexible deployment, high mobility and wide coverage are providing strong support for future communication networks. At present, Multi-UAV Relay-Assisted Communication has been widely used in many fields such as post-disaster emergency communications, remote monitoring, edge computing, and the Internet of Things. The successful implementation of these application scenarios depends on stable and reliable relay communication capabilities. Therefore, in a complex and dynamic environment, how to achieve efficient coordination, stable relay transmission and low-energy communication between multiple UAVs is one of the key research directions in this technology field.

[0003] To optimize the minimum user communication rate and the minimum UAV communication rate of multi-UAV relay-assisted communication, it is necessary to comprehensively consider the optimization of user group access, transmit power allocation, UAV trajectory and modulation order. Ignoring user grouping may cause some UAVs to be overloaded, resource utilization is unbalanced, and the overall network efficiency is reduced. Unoptimized transmit power will lead to channel mismatch, affecting the transmission rate and bit error rate. Insufficient trajectory optimization may lead to insufficient coverage or blind spots, affecting user communication services. Without adjusting the order, the spectrum utilization cannot be effectively improved. At present, the communication rate optimization of most multi-UAV relay-assisted communication systems is achieved by first optimizing user group access and then optimizing transmit power allocation and UAV trajectory planning through a two-step method. However, the two-step method cannot be flexibly adjusted according to the real-time demand of the UAV trajectory, which easily leads to resource waste or over-allocation. At the same time, due to the complexity of the optimization model, it is a major challenge to propose an efficient algorithm. Therefore, how to reduce the energy consumption of UAVs, improve user experience, and reduce the computational complexity of the algorithm under the premise of ensuring the communication rate is still an important problem and challenge to be solved in this field. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the shortcomings of the prior art, the present invention provides a multi-UAV relay-assisted communication rate optimization method based on MC-NOMA, which solves the problems raised in the above background technology.

[0006] (II) Technical solution

[0007] In order to achieve the above-mentioned purpose, the present invention specifically adopts the following technical solutions:

[0008] Multi-UAV Relay-Assisted Communication Rate Optimization Method Based on MC-NOMA, including the following steps:

[0009] S1: Establish an air-to-ground channel model and a power consumption model according to the characteristics of the multi-UAV assisted wireless communication scenario;

[0010] S2: Conduct problem modeling according to user access grouping constraints, physical constraints, and UAV performance constraints;

[0011] S3: Use the sequential optimization principle to decompose the originally multi-objective optimization problem with priorities into single-objective optimization problems;

[0012] S4: For the problem of maximizing the minimum user rate in different time slots, use the block coordinate descent principle to decompose the originally highly coupled mixed-integer non-linear programming problem into two sub-problems: user access grouping and power trajectory planning. Joint iterative optimization is performed on the two sub-problems. The user access grouping sub-problem is solved by introducing auxiliary variables to become a binary integer programming problem, and the power trajectory planning sub-problem is solved by introducing auxiliary variables and using the successive convex approximation method. Continuously cycle and iterate to optimize the two sub-problems until the convergence condition is reached;

[0013] S5: For the problem of maximizing the minimum communication rate between UAV base stations, optimize the UAV modulation problem based on the results of iterative optimization. The UAV modulation problem is solved by introducing auxiliary variables to become a binary integer programming problem.

[0014] Furthermore, in the S1 air-to-ground channel model, two scenarios in the air-to-ground channel model, line-of-sight (LoS) and non-line-of-sight (NLoS), are considered, and the channel gain considers the influence of the carrier frequency. The specific content is as follows: The path loss approximation is obtained by averaging the two cases through the occurrence probabilities of the two scenarios:

[0015]

[0016] The probability of LoS / NLoS and depends on the elevation angle of the UAV; The models of LoS / NLoS path fading and are expressed as follows:

[0017]

[0018] In this model, δ refers to the shadow fading, which is usually modeled as a log-normal distribution, where the mean of the log-normal distribution is μ and its variance σ 2 depends on the communication environment; represents the Rice distribution, and the Rice distribution is used to simulate small-scale fading such as multipath effects; Denote the free space propagation loss of the electromagnetic wave; then, the channel gain g between the UAV m and the user i m,i and the average path loss L m,i are converted as follows:

[0019]

[0020] Furthermore, the power consumption model in the S1 takes into account the transmission power of the UAV and the impact of the UAV trajectory on the energy consumption. The specific content is as follows: The energy consumption of the UAV mainly includes the flight energy consumption and the communication energy consumption of the UAV; the flight energy consumption consists of the hovering energy consumption and the energy consumption to overcome the wind resistance; assume that the average speed of the UAV m in the nth time slot is v m (n), and denote the mass of the UAV m as W m , then the flight energy consumption of the UAV is:

[0021]

[0022] where P 0 is the fixed power of the propeller; k is a constant; ρ is the air density; A is the disc area of the propeller; δ T is the duration of a time slice, and the total energy consumption of the UAV m is:

[0023]

[0024] where P m (n) is the sum of the power of the UAV m to each user and the base station in the nth time slot.

[0025] Furthermore, in the S2, the modulation method is added to the model and physical conditions such as the UAV flight speed and energy limit are considered. The optimization goal is to maximize the minimum user communication rate, which ensures fairness among users while maximizing the minimum communication rate between the UAV and the base station. In practical applications, the UAV is equipped with adaptive modulation and coding technology (AMC) to dynamically adjust the modulation order of the UAV communication to achieve high communication rates. The specific content is as follows: The quadrature amplitude modulation (QAM) technology is introduced into the communication between the UAV and the base station to maximize the performance of the assisted communication; therefore, in the time slot n, the communication rate between the UAV m and the ground station becomes as follows:

[0026]

[0027] In this case, U m,BS (n) represents the modulation order of the UAV m when sending information to the ground station in the nth time slot. In scenarios with high communication rate requirements, a high-order modulation method is generally adopted, so the modulation order of the UAV is given by the set U m,BS (n) ∈ {2, 4, 16, 64, 256}.

[0028] In each time slot, the initial velocity at the beginning of the time slot is used as the velocity within the time slot while considering physical limitations; the moving velocity v of UAV m m shall not exceed its maximum velocity Meanwhile, the maximum energy consumption E of UAV m m shall not exceed the maximum energy storage of the UAV

[0029] To ensure fairness among users, the optimization objective is designed to maximize the minimum user rate in different time slots while maximizing the minimum communication rate between UAV base stations; therefore, the multi-UAV relay-assisted communication objective function is constructed as follows:

[0030]

[0031] Furthermore, in step S3, the sequential optimization principle is applied to decompose the multi-objective optimization problem with priorities into single-objective optimization problems to be solved sequentially, and the introduction of the UAV base station communication rate constraint to avoid the situation of no feasible solution for maximizing the minimum communication rate between UAV base stations is specifically as follows: Since resources such as energy and frequency are shared during UAV-user communication and UAV base station communication, maximizing the minimum communication rate of users will result in an overly small solution space for maximizing the UAV base station communication rate or even being unable to meet the user upload service requirements. Therefore, it is set that the transmission rate of UAV m to the ground station exceeds the specified threshold R l , so a communication rate constraint is added when maximizing the minimum user rate in different time slots:

[0032]

[0033] In this way, the original multi-objective optimization problem is decomposed into the optimization problems:

[0034]

[0035] and:

[0036]

[0037] Furthermore, the specific content of solving the problem of maximizing the minimum user rate in different time slots by introducing auxiliary variables and using the block coordinate descent combined with the successive convex approximation method in step S4 is as follows: To study the user access grouping sub-problem, it is necessary to fix the transmit power of the relay UAV and the trajectory of the UAV. By maximizing the minimum channel gain from the user to the UAV, the channel condition of the mobile user with the worst ground channel condition is improved. By introducing the intermediate variable η, the user access grouping sub-problem can be simplified to:

[0038] min - η

[0039]

[0040] u = {0, 1}

[0041]

[0042] where g i,m [n] represents the channel gain magnitude between user i and UAV m in the nth time slot, and α represents the number of users that the UAV supports for access at the same time. This is a 0-1 integer programming problem, and the cvx toolbox Mosek optimization toolkit can be used to directly solve it to obtain the value of u i,m [n].

[0043] After obtaining the user access matrix, user grouping is performed. In the form of two users in a group, the users with the best and worst channel conditions are assigned to a group, and so on, to obtain the channel gain parameter matrix

[0044] To solve the UAV power trajectory optimization sub-problem, the user access grouping and pitch angle need to be fixed. After introducing the intermediate variable η, the problem is simplified as follows:

[0045] min - η

[0046]

[0047] 0 ≤ θ m,i ≤ 1

[0048] 0 ≤ θ m,BS ≤ 1

[0049]

[0050]

[0051] For the power trajectory optimization sub-problem, the successive convex approximation method is used to solve it iteratively. First, the non-convex constraint equation is transformed into a difference of convex (DC) form, and based on the difference of convex algorithm principle (DCA), the non-convex constraint is lower bounded and equivalently transformed to obtain a convex problem:

[0052] min - η

[0053]

[0054] where h 1 , h 2 , h 3 are the affine constraints obtained after lower bounding approximation, x m , y m , θ m,i are the decision variables, δ T , W m , H, Gm,BS is the parameter corresponding to the problem model. For convex problems, the cvx toolbox Mosek optimization toolkit can be used to directly solve for x m , y m , θ m,i values. Iterative solution is carried out based on the principle of successive convex approximation to solve the power trajectory sub-problem, and then iterative solution of the two sub-problems is carried out based on the principle of block coordinate descent.

[0055] After the user accesses the drone, user grouping is performed. In the form of two users in a group, the best and worst channel conditions are assigned to a group, and so on, to obtain the channel gain parameter matrix

[0056] For the modulation sub-problem, an auxiliary variable b m,1 , b m,2 , b m,3 , b m,4 , thus changing the U m,BS (n) variable from a discrete variable to a binary integer variable, and the original problem from a non-convex problem to a binary integer programming convex problem. The mathematical expression of the variable is as follows:

[0057] U m,BS (n) = b m,1 (n)4 + b m,2 (n)16 + b m,3 (n)64 + b m,4 (n)256.

[0058] Furthermore, the maximization of the minimum communication rate between UAV base stations in S5 is achieved by optimizing the modulation order. The specific content of changing the maximization of the minimum communication rate between UAV base stations to a binary integer programming by introducing auxiliary variables is: for the problem of maximizing the minimum communication rate between UAV base stations, auxiliary variables b m,1 (n), b m,2 (n), b m,3 (n), b m,4 (n) are introduced, thus changing the U m,BS (n) variable from a discrete variable to a binary integer variable, and the original problem from a non-convex problem to a binary integer programming convex problem. The mathematical expression of the variable is as follows:

[0059] U m,BS (n) = b m,1 (n)4 + b m,2 (n)16 + b m,3 (n)64 + b m,4 (n)256.

[0060] Similarly, for the introduction of the auxiliary variable η, the optimization problem becomes:

[0061] min-η

[0062]

[0063] b m,1 [n],b m,2 [n],b m,3 [n],b m,4 [n],b m,5 [n] ∈ {0, 1}

[0064] b m,1 [n] + b m,2 [n] + b m,3 [n] + b m,4 [n] + b m,5 [n] = 1

[0065] U m,BS [n] = 2b m,1 [n] + 4b m,2 [n] + 16b m,3 [n] + 64b m,4 [n] + 256b m,5 [n]

[0066] SNR m,BS [n] ≥ b m,1 [n]f 1 +b m,2 [n]f 2 +b m,3 [n]f 3 +b m,4 [n]f 4 +b m,5 [n]f 5

[0067] where f 1 ,f 2 ,f 3 ,f 4 ,B,SNR m,BS [n] is a constant calculated based on the result of the problem of maximizing the minimum user rate in different time slots. This is a 0-1 integer programming problem, and the cvx toolbox Mosek optimization toolkit can be directly used to solve it to obtain the value of U m .

[0068] Moreover, after user grouping, the situation of user access to the UAV changes, and the power obtained from the previous round of iterative optimization cannot match the new user grouping. Therefore, it is re-allocated according to the channel conditions of the users after the new user grouping;

[0069] Then, by constructing the non-convex constraint in the form of the difference of convex functions and then performing a lower bound approximation based on the difference of convex algorithm (DCA), the original non-convex power trajectory planning sub-problem is transformed into a convex problem, and the power trajectory planning sub-problem is solved iteratively. The specific method for the lower bound approximation of the difference of convex functions is as follows:

[0070] For the constraint:

[0071]

[0072] Where:

[0073]

[0074] This is a difference of convex constraint. Based on the DCA algorithm framework, a lower bound is used for approximation of h 1 to obtain:

[0075]

[0076] Where n 0 , B, D m,i are constants, is the result of the previous iteration. Substituting the result back into the original formula gives a convex constraint in the form of a convex function minus an affine function. For other non-convex constraints, the same method can be used for lower bound approximation to convert them into convex constraints.

[0077] (III) Beneficial Effects

[0078] Compared with the prior art, the present invention provides a method for optimizing the communication rate of multi-UAV relay-assisted communication based on MC-NOMA, having the following beneficial effects:

[0079] 1. The present invention provides an optimization model that is more in line with the actual scenario. The optimization objective is to maximize the minimum communication rate of users, effectively ensuring fairness among users while maximizing the minimum communication rate between UAV base stations, and taking into account physical conditions such as UAV flight energy consumption and UAV flight speed.

[0080] 2. The present invention incorporates the modulation method into the optimization model, and in the problem of maximizing the minimum communication rate between UAV base stations, it is transformed into a binary integer programming problem by introducing auxiliary variables, which can effectively improve the communication rate of UAV base stations and ensure that user information can be effectively transmitted back to the base station.

[0081] 3. The present invention avoids the problem of low resource utilization caused by static UAV access, and by adding user access grouping to the iterative process, it improves the resource utilization of UAVs.

[0082] 4. By means of the sequential optimization principle, the present invention decomposes the original multi-objective optimization problem into two single-objective optimization problems, effectively reducing the difficulty of problem solving and the computational efficiency of the algorithm.

[0083] 5. By using the block coordinate descent principle, the present invention decomposes the problem of maximizing the minimum user rate in different time slots with highly coupled variables into two sub-problems: user access grouping and power trajectory planning, and designs a joint optimization of the two sub-problems to solve the problems that the original planning problem is difficult to solve and the computational complexity is too high.

[0084] 6. By introducing auxiliary variables, the present invention changes the problem of maximizing the minimum communication rate between UAV base stations from an integer programming problem to a binary integer programming problem for solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 is a flowchart of the present invention;

[0086] Figure 2 is a schematic diagram of the scenario of the present invention;

[0087] Figure 3 is an optimization effect diagram of user access grouping of the present invention;

[0088] Figure 4 is an optimization effect diagram of the trajectory of the present invention;

[0089] Figure 5 is a comparison diagram of the minimum communication rate of users of the algorithm of the present invention in different scenarios;

[0090] Figure 6 is a comparison diagram of different algorithm schemes of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0091] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0092] Embodiment

[0093] As Figures 1-6 shown, a multi-UAV relay-assisted communication rate optimization method based on MC-NOMA proposed in an embodiment of the present invention includes the following steps:

[0094] S1: According to the multi-UAV-assisted wireless communication scenario such as Figure 2 , establish an air-to-ground channel model and a power consumption model;

[0095] Using the air-to-ground channel model, the ATG channel is divided into two scenarios: line-of-sight and non-line-of-sight. The path loss is obtained by averaging the two cases through their occurrence probabilities, which is described as follows:

[0096]

[0097] Probability of LoS / NLoS and depends on the elevation angle of the UAV. The models of LoS / NLoS path fading and are expressed as follows:

[0098]

[0099] In this model, δ refers to the shadow fading, which is usually modeled as a log-normal distribution, where the mean of the log-normal distribution is μ and its variance is σ 2 depends on the communication environment. represents the Rice distribution, which is used to simulate small-scale fading such as multipath effects. represents the free space propagation loss of electromagnetic waves. Then, the channel gain g m,i and the average path loss L m,i are converted as follows:

[0100]

[0101] The energy consumption of the UAV mainly includes the flight energy consumption and the communication energy consumption of the UAV. The flight energy consumption consists of the hovering energy consumption and the energy consumption to overcome wind resistance. Assume that the average speed of UAVm in the nth time slot is v m (n), and the mass of UAVm is denoted as W m , then the flight energy consumption of the UAV is:

[0102]

[0103] where P 0 is the fixed power of the propeller; k is a constant; ρ is the air density; A is the disc area of the propeller. The total energy consumption of UAV m is:

[0104]

[0105] where P m is the sum of the power of UAV m to each user and the base station.

[0106] S2: According to the user access grouping constraint, physical constraint, and UAV performance constraint, the problem is modeled;

[0107] Define u i,m(n ∈ {0, 1} is the user access parameter. When user i accesses UAV m, u i,m (n) = 1; otherwise, u i,m (n) = 0. Define σ i,j (n) ∈ {0, 1} as the channel gain parameter. When user i and user j belong to the same group and the channel gain of the jth user is better than that of the ith user otherwise, σ i,j (n) = 0.

[0108] The power allocation coefficients of UAV m to user i and the base station are θ m,i and θ m,BS . Then the communication rate of UAV m sending information to user i is:

[0109]

[0110] Introduce quadrature amplitude modulation (QAM) technology into the communication between the UAV and the base station to maximize the performance of the assisted communication. Therefore, in time slot n, the communication rate between UAV m and the ground station becomes as follows:

[0111]

[0112] In this case, U m,BS (n) represents the modulation order when UAV m sends information to the ground station in the nth time slot. In scenarios with high communication rate requirements, high-order modulation methods are generally used. Therefore, the modulation order of the UAV is given by the set U m,BS (n) ∈ {2, 4, 16, 64, 256}. Considering the physical constraints of the communication equipment and the performance constraints of the communication process, the total transmission power of the assisted UAV m shall not exceed its maximum transmission power The energy consumption of the assisted communication shall not exceed the maximum energy storage capacity of the assisted UAV It is expressed as follows:

[0113]

[0114] In each time slot, use the initial speed at the beginning of the time slot as the speed within the time slot, and the moving speed v of UAV m m shall not exceed its maximum speed It is expressed as follows:

[0115]

[0116] The calculation formula for the speed of UAV m in the nth time slot is:

[0117]

[0118] To ensure fairness among users, the optimization objective is designed to maximize the minimum user rate in different time slots while maximizing the minimum communication rate between UAV base stations. Therefore, the objective function of multi-UAV relay-assisted communication is constructed as follows:

[0119]

[0120] S3: Utilize the sequential optimization principle to decompose the multi-objective optimization problem with priorities into single-objective optimization problems;

[0121] Since UAV user communication and UAV base station communication share resources such as energy and frequency, maximizing the minimum communication rate of users will result in an overly small solution space for maximizing the communication rate of UAV base stations or even unable to meet the user upload service requirements. Therefore, it is set that the transmission rate from UAV m to the ground station exceeds the specified threshold R l Therefore, a communication rate constraint is added when maximizing the minimum user rate in different time slots:

[0122]

[0123] In this way, the original multi-objective optimization problem is decomposed into the following optimization problems:

[0124]

[0125] and:

[0126]

[0127] S4: For the problem of maximizing the minimum user rate in different time slots, use the block coordinate descent principle to decompose the originally highly coupled mixed-integer non-linear programming problem into two sub-problems: user access grouping and power trajectory planning. Joint iterative optimization is performed on the two sub-problems. The user access grouping sub-problem is solved by introducing auxiliary variables to transform it into a binary integer programming problem, and the power trajectory planning sub-problem is solved by introducing auxiliary variables and using the successive convex approximation method. The two sub-problems are continuously iteratively optimized until the convergence condition is reached;

[0128] To study the user access grouping sub-problem, it is necessary to fix the transmission power of the relay UAV and the trajectory of the UAV. By maximizing the minimum channel gain from the user to the UAV, the channel condition of the mobile user with the worst ground channel condition is improved. By introducing the intermediate variable η, the user access grouping sub-problem can be simplified as:

[0129] min - η

[0130]

[0131] where g i,m[n] represents the channel gain magnitude between user i and UAV m in the nth time slot, and α represents the number of users that the UAV can support for access at the same time. This is a 0-1 integer programming problem, which can be directly solved using the Mosek optimization toolkit in the cvx toolbox to obtain the value of u i,m [n]. The optimization effect is as Figure 3 shown.

[0132] After obtaining the user access matrix, user grouping is performed. In the form of pairwise grouping, the users with the best and worst channel conditions are assigned to a group, and so on, to obtain the channel gain parameter matrix

[0133] To solve the UAV power trajectory optimization sub-problem, it is necessary to fix the user access grouping and the pitch angle. After introducing the intermediate variable η, the problem is simplified as follows:

[0134] min - η

[0135]

[0136]

[0137] For the power trajectory optimization sub-problem, the method of successive convex approximation is used for iterative solution. First, the non-convex constraint equation is transformed into a difference of convex (DC) form, and the non-convex constraint is lower bounded and equivalently transformed into a convex problem based on the difference of convex algorithm principle (DCA):

[0138] min - η

[0139]

[0140] where h 1 , h 2 , h 3 are the affine constraints obtained after lower bounding approximation, x m , y m , θ m,i are decision variables, δ T , W m , H, G m,BS are the parameters corresponding to the problem model. The convex problem can be directly solved using the Mosek optimization toolkit in the cvx toolbox to obtain the values of x m , y m , θ m,i . Based on the principle of successive convex approximation, iterative solution is performed to solve the power trajectory sub-problem, and then based on the principle of block coordinate descent, iterative solution is performed for the two sub-problems. The trajectory optimization effect is as Figure 4 shown, Figure 5This is a comparison chart of the algorithm of the present invention under different numbers of drones, different energy storage limitations, and different drone flight speed limitations. Figure 6 This is a performance comparison chart of the algorithm of the present invention and the algorithms of other solutions in the same scenario.

[0141] S5: For the problem of maximizing the minimum communication rate between UAV base stations, the UAV modulation problem is optimized based on the iterative optimization results. The UAV modulation problem is transformed into a binary integer programming problem by introducing auxiliary variables. For the problem of maximizing the minimum communication rate between UAV base stations, auxiliary variable b m,1 (n), b m,2 (n), b m,3 (n), b m,4 (n), so that U m,BS (n) variables change from discrete variables to binary integer variables, and the original problem changes from a non-convex problem to a binary integer programming convex problem. The mathematical expressions of the variables are as follows:

[0142] U m,BS [n] = 2b m,1 [n] + 4b m,2 [n] + 16b m,3 [n] + 64b m,4 [n] + 256b m,5 [n]

[0143] At the same time, by introducing auxiliary variable η, the optimization problem becomes:

[0144] min - η

[0145]

[0146] b m,1 [n], b m,2 [n], b m,3 [n], b m,4 [n], b m,5 [n] ∈ {0, 1}

[0147] b m,1 [n] + b m,2 [n] + b m,3 [n] + b m,4 [n] + b m,5 [n] = 1

[0148] U m,BS [n] = 2b m,1 [n] + 4b m,2 [n] + 16b m,3 [n] + 64b m,4 [n] + 256b m,5 [n]

[0149] SNRm,BS [n] ≥ b m,1 [n]f 1 +b m,2 [n]f 2 +b m,3 [n]f 3 +b m,4 [n]f 4 +b m,5 [n]f 5

[0150] where f 1 ,f 2 ,f 3 ,f 4 ,B, SNR m,BS [n] is a constant calculated based on the result of maximizing the minimum user rate of different time slots. This is a 0-1 integer programming problem, which can be directly solved using the Mosek optimization toolbox of the cvx toolbox to obtain the value of U m .

[0151] Moreover, after user grouping, the situation of user access to the UAV changes, and the power obtained from the previous round of iterative optimization cannot match the new user grouping. Therefore, the power is reallocated according to the channel conditions of the users after the new user grouping;

[0152] Then, by constructing the non-convex constraint in the form of a convex difference and then performing a lower bound approximation based on the difference of convex algorithm (DCA), the original non-convex power trajectory planning sub-problem is transformed into a convex problem, and the power trajectory planning sub-problem is solved iteratively; the specific convex difference lower bound approximation method is as follows:

[0153] For the constraint:

[0154]

[0155] where:

[0156]

[0157] This is a convex difference constraint. Based on the DCA algorithm framework, a lower bound approximation is performed on h 1 to obtain:

[0158]

[0159] where n 0 , B, D m,i are constants, is the result of the previous round of iteration. Substituting the result back into the original formula gives a convex constraint in the form of a convex function minus an affine function. The same method can be used for other non-convex constraints to perform a lower bound approximation and convert them into convex constraints.

[0160] Based on the in - depth analysis of the problem modeling of multi - UAV relay - assisted wireless communication and the existing problems in current resource allocation and trajectory planning methods and algorithms, a resource allocation and trajectory planning method and algorithm specifically designed for multi - UAV relay - assisted wireless communication is successfully proposed. The emergence of this novel synchronization algorithm is of great value for further improving the performance of multi - UAV relay - assisted wireless communication systems.

[0161] The core innovation lies in that the MC - NOMA technology is carried on the relay UAV and the requirements for the modulation order in scenarios with high communication rate requirements are considered. The LOS / NLOS channel model in ATG is used in the process of problem modeling and the flight energy consumption of the UAV is considered in the power consumption model. In addition, the multi - objective optimization problem is decomposed into two single - objective optimization problems for sequential solution through sequential optimization. In the joint optimization algorithm, a joint optimization algorithm combining block coordinate descent and successive convex approximation methods is also designed to solve the problem of maximizing the minimum user rate in different time slots with highly coupled original variables. This design improvement enables the algorithm to effectively reduce the algorithm computational complexity of the system while improving the energy utilization efficiency and user experience of multi - UAV relay - assisted wireless communication systems.

[0162] Generally speaking, we propose a novel multi - UAV relay - assisted wireless communication model and design the corresponding solution algorithm. While innovatively constructing the channel and power consumption models, this model also considers the influence of the modulation order. By proposing an efficient solution algorithm, the performance and energy utilization rate of multi - UAV relay - assisted wireless communication systems are improved.

[0163] Finally, it should be noted that the above - mentioned are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A multi-UAV relay-assisted communication rate optimization method based on MC-NOMA, characterized by: The following steps are involved: S1: Establish air-to-ground channel model and power consumption model according to the characteristics of multi-UAV assisted wireless communication scenario; S2: Problem modeling based on user access grouping constraints, physical constraints, and drone performance constraints; S3: Use the sequential optimization principle to decompose the multi-objective optimization problem with priority into a single-objective optimization problem; S4: For the problem of maximizing the minimum user rate in different time slots, the block coordinate descent principle is used to decompose the original highly coupled mixed integer nonlinear programming problem into two sub-problems: user access grouping and power trajectory planning. The two sub-problems are jointly optimized iteratively. The user access grouping sub-problem is transformed into a binary integer programming problem by introducing auxiliary variables, and the power trajectory planning sub-problem is solved by introducing auxiliary variables and using the successive convex approximation method. The two sub-problems are continuously optimized iteratively to reach the convergence condition; S5: For the problem of maximizing the minimum communication rate between UAV base stations, the UAV modulation problem is optimized based on the results of iterative optimization. The UAV modulation problem is transformed into a binary integer programming solution by introducing auxiliary variables.

2. The multi-UAV relay-assisted communication rate optimization method based on MC-NOMA according to claim 1 is characterized in that: The air-to-ground channel model in S1 considers two scenarios in the air-to-ground channel model: line of sight (LoS) and non-line of sight (NLoS), and the channel gain considers the influence of the carrier frequency. The specific content is: the path loss approximation is obtained by averaging the two situations through the occurrence probability of the two scenarios: Probability of LoS / NLoS and Depends on the elevation angle of the drone; Model for LoS / NLoS path fading and It is expressed as follows: In this model, δ refers to the shadow fading, which is usually modeled as a log-normal distribution with mean μ and variance σ 2 Depends on the communication environment; Represents the Rice distribution, which is used to simulate small-scale fading such as multipath effects; represents the free space propagation loss of electromagnetic waves; then, the channel gain g between drone m and user i m,i and the average path loss L m,i The conversion is as follows:

3. The multi-UAV relay-assisted communication rate optimization method based on MC-NOMA according to claim 2 is characterized in that: The power consumption model in S1 takes into account the influence of the transmission power and trajectory of the UAV on the energy consumption. The specific contents are as follows: the energy consumption of the UAV mainly includes the flight energy consumption and communication energy consumption of the UAV; the flight energy consumption consists of the hovering energy consumption and the energy consumption of overcoming wind resistance; assuming that the average speed of UAV m in the nth time slot is v m (n), and the mass of UAVm is denoted as W m , then the UAV flight energy consumption is: Where P0 is the fixed power of the propeller; k is a constant; ρ is the air density; A is the propeller's rotating disk area; δ T is the duration of a time slice, and the total energy consumption of UAV m is: Where P m (n) is the sum of the powers delivered by UAV m to each user and the base station in the nth time slot.

4. The multi-UAV relay-assisted communication rate optimization method based on MC-NOMA according to claim 3 is characterized in that: In S2, the modulation mode is added to the model and physical conditions such as the flight speed and energy limitation of the UAV are taken into account. The optimization goal is to maximize the minimum user communication rate to ensure fairness among users and maximize the minimum communication rate between UAV base stations. The specific content is: in practical applications, the UAV is equipped with adaptive modulation and coding technology (AMC) to dynamically adjust the modulation order of UAV communication to achieve high communication rate. The specific content is: the orthogonal amplitude modulation (QAM) technology is introduced into the communication between the UAV and the base station to maximize the performance of auxiliary communication; therefore, in time slot n, the communication rate between UAV m and the ground station becomes as follows: In this case, U m,BS (n) represents the modulation order when UAV m sends information to the ground station in the nth time slot. In scenarios with high communication rate requirements, high-order modulation is generally used. Therefore, the modulation order of the UAV is composed of the set U m,BS (n)∈{2, 4, 16, 64, 256} is given; In each time slot, the initial speed at the beginning of the time slot is used as the speed within the time slot while considering physical limitations; the moving speed v of UAVm m Cannot exceed its maximum speed At the same time, the maximum energy consumption of UAV m is E m The maximum energy storage capacity of the drone cannot be exceeded In order to ensure fairness among users, the optimization objective is designed to maximize the minimum user rate in different time slots while maximizing the minimum communication rate between drone base stations; therefore, the multi-drone relay-assisted communication objective function is constructed as follows:

5. The multi-UAV relay-assisted communication rate optimization method based on MC-NOMA according to claim 4 is characterized in that: In S3, the sequential optimization principle is applied to decompose the multi-objective optimization problem with priority into single-objective optimization problems to be solved in sequence, and the UAV base station communication rate constraint is introduced to avoid the situation where there is no feasible solution to maximize the minimum UAV base station communication rate. The specific content is: since the communication between UAV users and the communication between UAV base stations share energy, frequency and other resources, when maximizing the user's minimum communication rate, the solution space for maximizing the UAV base station communication rate will be too small and may even fail to meet the user's upload service requirements. Therefore, the transmission rate from UAV m to the ground station is set to exceed the specified threshold R l , so the communication rate constraint is added when maximizing the minimum user rate in different time slots: In this way, the original multi-objective optimization problem is decomposed into an optimization problem: and:

6. The multi-UAV relay-assisted communication rate optimization method based on MC-NOMA according to claim 5 is characterized in that: The specific content of the problem of maximizing the minimum user rate in different time slots in S4 is solved by introducing auxiliary variables through block coordinate descent combined with successive convex approximation method: In order to study the user access grouping sub-problem, it is necessary to fix the transmission power of the relay drone and the trajectory of the drone, and improve the channel condition of the mobile user with the worst ground channel condition by maximizing the minimum channel gain from the user to the drone. The user access grouping sub-problem can be simplified by introducing the intermediate variable η: min-η u＝{0,1} where g i,m [n] represents the channel gain between user i and drone m in the nth time slot, and α represents the number of users that the drone supports accessing at the same time. This is a 0-1 integer programming problem, which can be solved directly using the Mosek optimization toolkit in the cvx toolbox to obtain u i,m The value of [n]; After obtaining the user access matrix, users are grouped in pairs, with the best and worst channel conditions assigned to one group, and so on, to obtain the channel gain parameter matrix To solve the UAV power trajectory optimization subproblem, it is necessary to fix the user access group and pitch angle. After introducing the intermediate variable η, the problem is simplified as follows: min-η 0≤θ m,i ≤1 0≤θ m,BS ≤1 The power trajectory optimization subproblem is solved iteratively using the successive convex approximation method. First, the non-convex constraint equation is transformed into a convex difference (DC) form. Based on the convex difference algorithm (DCA), the non-convex constraint is approximated by the lower bound and equivalently transformed to obtain a convex problem: min-η Among them, h1, h2, and h3 are the affine constraints obtained after the lower bound approximation, and x m ,y m ,θ m,i is the decision variable, δ T , W m , H.G. m,BS is the parameter corresponding to the problem model. For convex problems, the Mosek optimization toolkit in the cvx toolbox can be used to directly solve them to obtain x m ,y m ,θ m,i The power trajectory subproblem is solved iteratively based on the principle of successive convex approximation, and then the two subproblems are solved iteratively based on the block coordinate descent principle.

7. The multi-UAV relay-assisted communication rate optimization method based on MC-NOMA according to claim 6 is characterized in that: In S5, the maximization of the minimum UAV base station communication rate is achieved by optimizing the modulation order. The maximization of the minimum UAV base station communication rate is converted into a binary integer programming solution by introducing auxiliary variables. The specific content is: for the maximization of the minimum UAV base station communication rate problem, the auxiliary variable b is introduced. m,1 (n),b m,2 (n),b m,3 (n),b m,4 (n), so that U m,BS (n) The variables are changed from discrete variables to binary integer variables. The original problem is changed from a non-convex problem to a binary integer programming convex problem. The mathematical expression of the variables is as follows: U m,BS [n]=2b m,1 [n]+4b m,2 [n]+16b m,3 [n]+64b m,4 [n]+256b m,5 [n] At the same time, the auxiliary variable η is introduced, and the optimization problem becomes: min-η b m,1 [n],b m,2 [n],b m,3 [n],b m,4 [n],b m,5 [n]∈{0,1} b m,1 [n]+b m,2 [n]+b m,3 [n]+b m,4 [n]+b m,5 [n]=1 U m,BS [n]=2b m,1 [n]+4b m,2 [n]+16b m,3 [n]+64b m,4 [n]+256b m,5 [n] SNR m,BS [n]≥b m,1 [n]f1+b m,2 [n]f2+b m,3 [n]f3+b m,4 [n]f4+b m,5 [n]f5 where f1, f2, f3, f4, B, SNR m,BS [n] is a constant calculated based on the result of the problem of maximizing the minimum user rate for different time slots. This is a 0-1 integer programming problem that can be solved directly using the Mosek optimization toolkit in the cvx toolbox to obtain U m The value of .

Images