Multi-unmanned aerial vehicle assisted wireless emergency communication optimization method

Abstract

The invention discloses a multi-unmanned aerial vehicle auxiliary wireless emergency communication optimization method, and belongs to the technical field of communication. According to the invention, on the basis of downlink communication energy budget of each unmanned aerial vehicle, a system throughput maximization problem of a multi-unmanned aerial vehicle assisted wireless emergency communication system is modeled into a mathematical optimization problem under multiple constraints; the method comprises the following steps: decomposing an original non-convexity optimization problem into three sub-problems of user association strategy optimization of an unmanned aerial vehicle and a ground user, unmanned aerial vehicle transmitting power strategy optimization and unmanned aerial vehicle trajectory optimization based on a block coordinate descent idea; a non-convex sub-problem is converted into a convex problem which can be solved through a corresponding convex optimization tool by introducing auxiliary variables, continuously convex approximation and other mathematical means, and the three sub-problems are iteratively solved through an alternate optimization mode to obtain an approximate optimal solution of an original problem. According to the invention, accurate management of communication energy can be realized, and the capacity of the system is effectively improved.

Description

Technical Field

[0001] The present invention belongs to the field of communication technologies, and particularly relates to an optimization method for multi-UAV-assisted wireless emergency communication. Background Art

[0002] Supported by B5G and 6G communication technologies, the space-air-ground integrated network (SAGIN), as an emerging network architecture, can achieve seamless coverage and reliable transmission for global applications by deeply integrating space-based networks, satellite-based networks, and terrestrial cellular networks across systems, technologies, and applications. However, in remote or disaster-affected areas, relying solely on terrestrial networks cannot ensure reliable wireless services. Due to their high mobility and easy deployment, etc., unmanned aerial vehicles (UAVs) can adapt to extreme environments and complex geographical conditions, and have become an important hub connecting satellite-based networks and terrestrial networks. Their three-dimensional mobility can significantly increase the line-of-sight connection probability and quality of service (QoS) of ground users, providing a flexible and efficient solution for network coverage and communication support.

[0003] However, the flight time of UAVs is limited by the battery capacity, which directly affects their ability to provide continuous services in the air, and the high-power communication requirements, long flights, and the energy required for device operation exacerbate this problem. In addition, to maintain a stable connection with ground users, UAVs need to perform frequent signal transmissions and relaying, which will quickly consume their communication energy.

[0004] In summary, the contradiction between the limited communication and flight resources of UAV-assisted communication systems and the diverse wireless service demands of users is becoming increasingly prominent, and it is necessary to study the trajectories and resource allocation of UAV-assisted communication systems. Summary of the Invention

[0005] To solve the above problems, the present invention proposes an optimization method for multi-UAV-assisted wireless emergency communication. This method is implemented based on mathematical methods such as block coordinate descent and successive convex approximation, and can jointly optimize the flight trajectories, transmission powers, and user association strategies of UAVs while finely managing the downlink communication energy of multi-UAV systems, maximizing the minimum throughput of the system and improving the performance of UAV-assisted wireless emergency communication systems.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] An optimization method for multi-UAV-assisted wireless emergency communication, used to optimize the throughput of a multi-UAV-assisted wireless emergency communication system, includes the following steps:

[0008] Step S1: construct an optimization problem of maximizing the minimum system throughput under multiple constraints in a multi-UAV assisted wireless emergency communication system, and decompose the problem into three sub-problems according to the block coordinate descent method, wherein the first sub-problem is to solve the optimal user association strategy of the UAV under a fixed UAV trajectory and transmission power strategy, the second sub-problem is to solve the optimal transmission power of the UAV under a fixed UAV trajectory and user association strategy, and the third sub-problem is to solve the optimal trajectory under a fixed UAV power transmission strategy and user association strategy;

[0009] Step S2: Based on the initial trajectory and power of the drone, solve the first sub-problem and obtain the user association strategy ;

[0010] Step S3: Based on user association strategy and the initial trajectory of the UAV, solve the second sub-problem and obtain the UAV power transmission strategy ;

[0011] Step S4: Based on user association strategy and UAV power transmission strategy , solve the third sub-problem and get the trajectory of the drone ;

[0012] Step S5: Repeat steps S2, S3, and S4 to alternately solve the three sub-problems to achieve continuous iterative optimization of the system throughput. When the preset conditions are met, the optimal user association strategy is returned. , Optimal UAV launch power strategy , Optimal UAV flight trajectory .

[0013] Furthermore, the multi-UAV assisted wireless emergency communication system is A drone is used as an aerial base station to provide Providing communication services to individual users, The drone is connected to the core network through a low-orbit satellite. In the process, the UAV moves in a circle at a fixed height, the position of the ground user is known and unchanged, and the mission time is divided into time slots, and the length of each time slot is , the position of the UAV remains approximately unchanged in each time slot.

[0014] Furthermore, in step S1, the optimization problem of maximizing the minimum system throughput under multiple constraints in the multi-UAV assisted wireless emergency communication system is constructed, and the specific method is as follows:

[0015] The path between the UAV and the ground user is modeled as a free space path loss model. With ground users The channel gain between is:

[0016] h u , k [ n ] = β 0 d u , k − 2 [ n ]

[0017] in, For the reference distance The channel power gain at d u , k [ n ] For drones With ground users The distance between the two, when the drone is at a fixed altitude, In each time slot, the drone With ground users The distance between is expressed as:

[0018] d u , k [ n ] = ‖ q u [ n ] − u k [ n ] ‖ 2 + H 2

[0019] in, q u [ n ] For drones In time slot location, u k [ n ] For users In time slot location, is the flight altitude of the drone. According to Shannon's theorem, the drone With ground node In time slot The transmission rate within is:

[0020] R u , k [ n ] = W l o g 2 ( 1 + p u [ n ] h u , k [ n ] s 2 ) = W log 2 ( 1 + c 0 p u [ n ] ‖ q u [ n ] − u k [ n ] ‖ 2 + H 2 )

[0021] in, is the system bandwidth, p u [ n ] For drones In time slot The transmission power, Indicates the reference distance The signal-to-noise ratio at is the Gaussian white noise power;

[0022] In any time slot, each drone base station provides services to at most one ground user. At the same time, in any time slot, each ground user can only communicate with one drone base station. b u , k [ n ] To represent the drone base station With ground users The relationship between them; when the drone base station With ground users When establishing a connection, there is b u , k [ n ] = 1 ,otherwise b u , k [ n ] = 0 ;

[0023] During the service period of the drone base station, the system throughput is expressed as:

[0024] R total = ∑ n = 1 N ∑ u = 1 U ∑ k = 1 K b u , k [ n ] R u , k [ n ] t

[0025] in, is the time slot length;

[0026] As an aerial communication platform, the drone base station provides wireless services to ground users. The communication energy consumption of a drone base station during the entire service cycle is:

[0027] E total = ∑ n = 1 N ∑ u = 1 U p u [ n ] t

[0028] Introducing auxiliary variables represents the minimum throughput of the system, and the system throughput maximization problem based on the UAV communication energy budget is modeled as:

[0029] max S , X , Q , or or st { C 1 : ∑ n = 1 N ∑ u = 1 U ∑ k = 1 K t b u , k [ n ] R u , k [ n ] ≥ or C 2 : b u , k [ n ] = { 0 , 1 } , ∀ u ∈ U , k ∈ K , n ∈ N C 3 : ∑ u = 1 U b u , k [ n ] ≤ 1 , ∀ k ∈ K C 4 : ∑ k = 1 K b u , k [ n ] ≤ 1 , ∀ u ∈ U C 5 : 0 ≤ p u [ n ] ≤ p max , ∀ u ∈ U , n ∈ N C 6 : ∑ n = 1 N ∑ u = 1 U p u [ n ] t ≤ E e , ∀ n ∈ N C 7 : ‖ q u [ n + 1 ] − q u [ n ] ‖ ≤ v m a x t , ∀ u ∈ U , n ∈ N C 8 : ‖ q u [ n ] − q m [ n ] ‖ 2 ≥ ( d s ) 2 , ∀ u , m ∈ U , n ∈ N , u ≠ m C 9 : q u [ 1 ] = q u [ N ] , ∀ u ∈ U , n ∈ N

[0030] in, , and are user association, UAV transmission power and UAV air position set respectively; C2-C4 are user association constraints; is the maximum transmission power of the UAV, is the communication energy budget of the UAV, C5 and C6 are the transmission power and information transmission energy constraints of the UAV; is the maximum flight speed of the drone, is the anti-collision distance between UAVs, C7-C9 are the trajectory constraints of the UAV base station, where C7 constrains the speed of the same UAV between adjacent time slots, C8 constrains the safety distance between different UAVs, and C9 is a circular trajectory constraint to ensure that the UAV base station returns to the initial position in the end.

[0031] Furthermore, in step S1, the first sub-problem is expressed as:

[0032] max X , Q , or or st { C 1 : ∑ n = 1 N ∑ u = 1 U ∑ k = 1 K t b u , k [ n ] R u , k [ n ] ≥ or C 2 : b u , k [ n ] = { 0 , 1 } , ∀ u ∈ U , k ∈ K , n ∈ N C 3 : ∑ u = 1 U b m , k [ n ] ≤ 1 , ∀ k ∈ K , n ∈ N C 4 : ∑ k = 1 K b u , k [ n ] ≤ 1 , ∀ u ∈ U , n ∈ N

[0033] The second sub-problem is expressed as:

[0034] max S , Q , or or st { C 1 : ∑ n = 1 N ∑ u = 1 U ∑ k = 1 K t b u , k [ n ] R u , k ( { S } { Q } ) ≥ or C 5 : 0 ≤ p u [ n ] ≤ p max , ∀ u ∈ U , n ∈ N C 6 : ∑ n = 1 N ∑ u = 1 U p u [ n ] t ≤ E e , ∀ n ∈ N

[0035] The third sub-problem is expressed as:

[0036] max S , X , or or st { C 1 : ∑ n = 1 N ∑ u = 1 U ∑ k = 1 K t b u , k [ n ] R u , k [ n ] ≥ or C 7 : ‖ q u [ n + 1 ] − q u [ n ] ‖ ≤ v m a x t , ∀ u ∈ U , n ∈ N C 8 : ‖ q u [ n ] − q m [ n ] ‖ 2 ≥ ( d s ) 2 , ∀ u , m ∈ U , n ∈ N , u ≠ m C 9 : q u [ 1 ] = q u [ N ] , ∀ u ∈ U , n ∈ N .

[0037] Furthermore, the specific method of step S2 is:

[0038] Binary variables b u , k [ n ] Relaxation is a continuous variable, then the first subproblem is expressed as:

[0039] max X , Q , or or st { C 1 : ∑ n = 1 N ∑ u = 1 U ∑ k = 1 K t b u , k [ n ] R u , k [ n ] ≥ or C 2 : 0 ≤ b u , k [ n ] ≤ 1 , ∀ u ∈ U , k ∈ K , n ∈ N C 3 : ∑ u = 1 U b m , k [ n ] ≤ 1 , ∀ k ∈ K , n ∈ N C 4 : ∑ k = 1 K b u , k [ n ] ≤ 1 , ∀ u ∈ U , n ∈ N

[0040] This problem is a standard linear programming problem. It is solved using the mathematical tools of linear programming and the optimization toolkit to obtain the user's associated solution. .

[0041] Furthermore, in step S3, the second sub-problem is a standard linear programming problem, which is solved using the mathematical tools of linear programming and the optimization toolkit to obtain the UAV base station power transmission strategy. .

[0042] Furthermore, the specific method of step S4 is:

[0043] In the third subproblem, C1 and C8 are non-convex constraints;

[0044] For the non-convex constraint C1, Lieutenant General ‖ q u [ n ] − u k [ n ] ‖ 2 As a whole, R u , k [ n ] About ‖ q u [ n ] − u k [ n ] ‖ 2 A quasi-convex function, given any local point ‖ q j [ n ] − u k [ n ] ‖ 2 By first-order Taylor expansion and sorting, we can get R u , k [ n ] The lower bound R ¯ l b [ n ] :

[0045] R ¯ l b [ n ] = W ( α j [ n ] − β j [ n ] ( ‖ q [ n ] − u k [ n ] ‖ 2 − ‖ q j [ n ] − u k [ n ] ‖ 2 ) )

[0046] in:

[0047] α j [ n ] = l o g 2 ( 1 + c 0 p u , k [ n ] ‖ q j [ n ] − u k [ n ] ‖ 2 ) β j [ n ] = ( l o g 2 e ) c 0 ( H 2 + c 0 p u , k [ n ] + ‖ q j [ n ] − u k [ n ] ‖ 2 ) ( H 2 + ‖ q j [ n ] − u k [ n ] ‖ 2 )

[0048] Then C1 is expressed as:

[0049] ∑ n = 1 N ∑ m = 1 M b m [ n ] R ¯ l b [ n ] t ≥ or

[0050] Put C8 in q u j [ n ] and q m j [ n ] Perform a first-order Taylor expansion at ‖ q u [ n ] − q m [ n ] ‖ 2 The global lower bound of :

[0051] ‖ q u [ n ] − q m [ n ] ‖ 2 ≥ 2 ( q u j [ n ] − q m j [ n ] ) T × ( q u [ n ] − q m [ n ] )  − ‖ q u j [ n ] − q m j [ n ] ‖ 2

[0052] Thus, C8 is transformed into a linear constraint;

[0053] At this point, the third sub-problem is expressed as:

[0054] max S , X , or or st { C 1 : ∑ n = 1 N ∑ m = 1 M b m [ n ] R ¯ l b [ n ] t ≥ or C 2 : ‖ q u [ n + 1 ] − q u [ n ] ‖ ≤ v m a x t , ∀ u ∈ U , n ∈ N C 3 : 2 ( q u j [ n ] − q m j [ n ] ) T × ( q u [ n ] − q m [ n ] ) − ‖ q u j [ n ] − q m j [ n ] ‖ 2 ≥ ( d s ) 2 C 4 : q u [ 1 ] = q u [ N ] , ∀ u ∈ U , n ∈ N

[0055] This problem is a standard convex optimization problem. The interior point method is used to solve it and the trajectory of the drone is obtained. .

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] (1) Based on the consideration of the energy consumption of downlink communication of drones in the integrated air-ground-space network, this invention proposes an iterative algorithm to perform system resource allocation and drone trajectory optimization. With the goal of maximizing the minimum throughput of the system, a problem of jointly optimizing the user association, transmission power and flight trajectory of the drone is established. There is a nonlinear coupling relationship between the optimization variables in this problem. It is a non-convex optimization problem and is difficult to solve directly. The original problem is decomposed into three sub-problems through the block coordinate descent method. Based on the mathematical characteristics of each sub-problem, the non-convex problem is transformed into a convex problem by introducing slack variables and continuous convex approximation. By alternately iteratively optimizing the three sub-problems, an approximate suboptimal solution to the original non-convex problem is obtained.

[0058] (2) Under the alternating optimization algorithm proposed in the present invention, the UAV adopts a more reasonable flight trajectory by selectively providing associated services to ground users and selecting the transmission power under a strict energy budget, which effectively improves the system throughput and shows obvious performance advantages compared with other comparison algorithms.

[0059] In summary, the present invention utilizes the strong maneuverability and flexible deployment of drones and the advantages of collaborative ground communication between multiple drones. Through the fine management of drone communication resources, the full utilization of airspace resources and theoretical analysis and derivation, an alternating optimization algorithm is adopted to realize the joint optimization of drone trajectory, transmission power and user association strategy, thereby achieving maximum throughput transmission of the emergency system, improving the performance of the emergency communication system, and can be used for high-performance downlink transmission of multi-drone assisted wireless emergency communications in an integrated air-ground-space network. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 Schematic diagram of a UAV-assisted wireless emergency communication system in an embodiment of the present invention.

[0061] Figure 2 It is a flow chart of a multi-UAV assisted wireless emergency communication optimization method in an embodiment of the present invention.

[0062] Figure 3 Schematic diagram of the initial trajectory simulation result of the UAV in an embodiment of the present invention.

[0063] Figure 4 Schematic diagram of simulation results of the user association strategy of drone 1 in an embodiment of the present invention.

[0064] Figure 5 Schematic diagram of simulation results of the user association strategy of drone 2 in an embodiment of the present invention.

[0065] Figure 6 Schematic diagram of the simulation results of the UAV transmission power strategy in an embodiment of the present invention.

[0066] Figure 7 Schematic diagram of the simulation result of the optimized trajectory of the UAV in the embodiment of the present invention.

[0067] Figure 8 2 is a schematic diagram of the simulation results of the system throughput at different deployment altitudes of the UAV in an embodiment of the present invention.

[0068] Figure 9 4 is a schematic diagram of the simulation results of the system throughput of the UAV under different mission cycles in an embodiment of the present invention.

[0069] Figure 10 It is a schematic diagram of simulation results of the variation of system throughput with the number of iterations under different algorithms in an embodiment of the present invention.

[0070] Figure 11 4 is a schematic diagram of the simulation results of system throughput under different energy budgets of a UAV in an embodiment of the present invention. DETAILED DESCRIPTION

[0071] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0072] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, "plurality" means two or more, unless otherwise expressly specified.

[0073] A multi-UAV assisted wireless emergency communication optimization method is applied to a UAV assisted emergency communication system in an air-ground integrated network. Figure 1 As shown, the system is deployed A drone is used as an aerial base station to provide ground It provides communication services to each user and establishes a connection with the core network through low-orbit satellites. In the time, the UAV base station moves in a circle at a fixed height, and the position of the ground user is known and unchanged. In order to facilitate the description of the position of the UAV base station, the mission time is divided into time slots, where the length of each time slot is , the position of the drone remains approximately unchanged in each time slot. In order to avoid interference between users, the frequency division multiple access (FDMA) technology is used.

[0074] When a UAV communicates with a ground user at a certain altitude, there is a line-of-sight transmission path most of the time. Therefore, in this embodiment, the path between the UAV and the ground user is modeled as a free space path loss model. With ground users The channel gain between is:

[0075] h u , k [ n ] = β 0 d u , k − 2 [ n ]

[0076] in For the reference distance The channel power gain at is affected by the system carrier frequency and the antenna gain. d u , k [ n ] For drones With ground users The distance between the two, when the drone is at a fixed altitude, In each time slot, the UAV and the ground user The distance between them can be expressed as:

[0077] d u , k [ n ] = ‖ q u [ n ] − u k [ n ] ‖ 2 + H 2

[0078] in q u [ n ] For drones In time slot location, u k [ n ] For users Time slot location, is the flight altitude of the drone. According to Shannon's theorem, the drone With ground node In time slot The transmission rate within is:

[0079] R u , k [ n ] = W l o g 2 ( 1 + p u [ n ] h u , k [ n ] s 2 ) = W log 2 ( 1 + c 0 p u [ n ] ‖ q u [ n ] − u k [ n ] ‖ 2 + H 2 )

[0080] in is the system bandwidth, p u [ n ] For drones In time slot The transmission power, is the noise power spectral density, Indicates the reference distance The signal-to-noise ratio at .

[0081] In any time slot, each drone base station provides services to at most one ground user. At the same time, each ground user can only communicate with one drone base station in any time slot. For ease of representation, a binary variable is introduced b u , k [ n ] To represent the drone base station With ground users When the drone base station With ground users To establish a connection, b u , k [ n ] = 1 ;on the contrary b u , k [ n ] = 0 . Then, during the service period of the drone base station, the system throughput can be expressed as:

[0082] R total = ∑ n = 1 N ∑ u = 1 U ∑ k = 1 K b u , k [ n ] R u , k [ n ] t

[0083] As an aerial communication platform, the UAV base station provides wireless services to ground users. Its communication energy consumption is much less than the flight energy consumption of the UAV, but it determines the service quality for ground users. During the entire service time period, The communication energy consumption of a drone base station during the entire service cycle is:

[0084] E total = ∑ n = 1 N ∑ u = 1 U p u [ n ] t

[0085] This method aims to maximize the minimum throughput of the multi-UAV-assisted wireless emergency communication system under limited communication energy by jointly optimizing the association strategy between the UAV base station and ground users, the UAV base station trajectory and the transmission power.

[0086] like Figure 2 As shown, the method comprises the following steps:

[0087] Step S1: Since the multivariable coupled UAV base station trajectory and resource allocation problem is non-convex, in order to facilitate the solution, an auxiliary variable is introduced Represents the minimum throughput of the system, then the system throughput maximization problem based on the UAV communication energy budget can be modeled as:

[0088] max S , X , Q , or or st { C 1 : ∑ n = 1 N ∑ u = 1 U ∑ k = 1 K t b u , k [ n ] R u , k ( { S } { X } { Q } ) ≥ or C 2 : b u , k [ n ] = { 0 , 1 } , ∀ u ∈ U , k ∈ K , n ∈ N C 3 : ∑ u = 1 U b u , k [ n ] ≤ 1 , ∀ k ∈ K C 4 : ∑ k = 1 K b u , k [ n ] ≤ 1 , ∀ u ∈ U C 5 : 0 ≤ p u [ n ] ≤ p max , ∀ u ∈ U , n ∈ N C 6 : ∑ n = 1 N ∑ u = 1 U p u [ n ] t ≤ E e , ∀ n ∈ N C 7 : ‖ q u [ n + 1 ] − q u [ n ] ‖ ≤ v m a x t , ∀ u ∈ U , n ∈ N C 8 : ‖ q u [ n ] − q m [ n ] ‖ 2 ≥ ( d s ) 2 , ∀ u , i ∈ U , n ∈ N , u ≠ m C 9 : q u [ 1 ] = q u [ N ] , ∀ u ∈ U , n ∈ N

[0089] in, , and They are user association, drone transmission power and drone air position set, C2-C4 are user association constraints, C5 and C6 are drone transmission power and information transmission energy constraints, C7-C9 are drone base station trajectory constraints, among which C7 constrains the speed of the same drone between adjacent time slots, C8 constrains the safety distance between different drones, and C9 is a circular trajectory constraint to ensure that the drone base station returns to the initial position in the end. Through analysis, it can be seen that since C1 contains coupling terms of power and trajectory variables, the constructed optimization problem is non-convex and difficult to solve directly.

[0090] Block coordinate descent is an iterative optimization algorithm for solving multivariable optimization problems. The principle is to divide the variables in the optimization problem into several groups (or blocks), and then in each iteration, only update the variables of one block, while keeping the variables of other blocks unchanged or using the previous values. In this way, the entire optimization problem is decomposed into multiple sub-problems, each of which only involves the variables of one block, thereby simplifying the solution of the optimization problem.

[0091] Decompose the original problem into subproblems according to the block coordinate descent method (Solving the optimal user association strategy for drones under fixed drone trajectories and transmission power), its mathematical optimization model can be expressed as:

[0092] max X , Q , or or st { C 1 : ∑ n = 1 N ∑ u = 1 U ∑ k = 1 K t b u , k [ n ] R u , k ( { X } { Q } ) ≥ or C 2 : b u , k [ n ] = { 0 , 1 } , ∀ u ∈ U , k ∈ K , n ∈ N C 3 : ∑ u = 1 U b m , k [ n ] ≤ 1 , ∀ k ∈ K , n ∈ N C 4 : ∑ k = 1 K b u , k [ n ] ≤ 1 , ∀ u ∈ U , n ∈ N

[0093] Sub-questions (Solution of the optimal power strategy of drones under fixed drone trajectories and user association strategies), its mathematical optimization model can be expressed as:

[0094] max S , Q , or or st { C 1 : ∑ n = 1 N ∑ u = 1 U ∑ k = 1 K t b u , k [ n ] R u , k ( { S } { Q } ) ≥ or C 5 : 0 ≤ p u [ n ] ≤ p max , ∀ u ∈ U , n ∈ N C 6 : ∑ n = 1 N ∑ u = 1 U p u [ n ] t ≤ E e , ∀ n ∈ N

[0095] Sub-questions (Solution of the optimal flight trajectory of a drone under the drone fixed power transmission strategy and user association strategy), its mathematical optimization model can be expressed as:

[0096] max S , X , or or st { C 1 : ∑ n = 1 N ∑ u = 1 U ∑ k = 1 K t b u , k [ n ] R u , k ( { S } { X } ) ≥ or C 7 : ‖ q u [ n + 1 ] − q u [ n ] ‖ ≤ v m a x t , ∀ u ∈ U , n ∈ N C 8 : ‖ q u [ n ] − q m [ n ] ‖ 2 ≥ ( d s ) 2 , ∀ u , i ∈ U , n ∈ N , u ≠ m C 9 : q u [ 1 ] = q u [ N ] , ∀ u ∈ U , n ∈ N

[0097] Step S2: Solve subproblems based on initial drone base station trajectory and power , get the user association strategy , when the flight trajectory of the drone and transmit power When determining, first transform the binary variable b u , k [ n ] Relaxed to continuous variables, then the subproblem It is expressed as:

[0098] max X , Q , or or st { C 1 : ∑ n = 1 N ∑ u = 1 U ∑ k = 1 K t b u , k [ n ] R u , k ( { X } { Q } ) ≥ or C 2 : 0 ≤ b u , k [ n ] ≤ 1 , ∀ u ∈ U , k ∈ K , n ∈ N C 3 : ∑ u = 1 U b m , k [ n ] ≤ 1 , ∀ k ∈ K , n ∈ N C 4 : ∑ k = 1 K b u , k [ n ] ≤ 1 , ∀ u ∈ U , n ∈ N

[0099] From the above formula, we can see that the sub-problem This is a standard linear programming problem and can be solved using linear programming mathematical tools and optimization toolkits.

[0100] Step S3: Based on user association and the initial UAV base station trajectory solving subproblem , get the UAV base station power transmission strategy Sub-problem This is a standard linear programming problem and can also be solved using linear programming mathematical tools and optimization toolkits.

[0101] Step S4: In the sub-problem In , C1 and C8 are non-convex constraints. First, define the function:

[0102]

[0103] Then its first-order derivative is:

[0104]

[0105] The corresponding second-order derivative is:

[0106]

[0107] when and hour, but is a convex function. At any point Performing a first-order Taylor expansion on it gives the global lower bound of the original function. , and there are:

[0108]

[0109] Based on this idea, Lieutenant General ‖ q u [ n ] − u k [ n ] ‖ 2 As a whole, R u , k [ n ] About ‖ q u [ n ] − u k [ n ] ‖ 2 A quasi-convex function, given any local point ‖ q j [ n ] − u k [ n ] ‖ 2 By first-order Taylor expansion and sorting, we can get R u , k [ n ] The lower bound R ¯ l b [ n ] :

[0110] R ¯ l b [ n ] = W ( α j [ n ] − β j [ n ] ( ‖ q [ n ] − u k [ n ] ‖ 2 − ‖ q j [ n ] − u k [ n ] ‖ 2 ) )

[0111] in:

[0112] α j [ n ] = l o g 2 ( 1 + c 0 p u , k [ n ] ‖ q j [ n ] − u k [ n ] ‖ 2 ) β j [ n ] = ( l o g 2 e ) c 0 ( H 2 + c 0 p u , k [ n ] + ‖ q j [ n ] − u k [ n ] ‖ 2 ) ( H 2 + ‖ q j [ n ] − u k [ n ] ‖ 2 )

[0113] Then C1 can be expressed as:

[0114] ∑ n = 1 N ∑ m = 1 M b m [ n ] R ¯ l b [ n ] t ≥ or

[0115] Similarly, put C8 in q u j [ n ] and q m j [ n ] Perform a first-order Taylor expansion at:

[0116] ‖ q u [ n ] − q m [ n ] ‖ 2 ≥ 2 ( q u j [ n ] − q m j [ n ] ) T × ( q u [ n ] − q m [ n ] )  − ‖ q u j [ n ] − q m j [ n ] ‖ 2

[0117] This transforms C8 into a linear constraint. can be re-expressed as:

[0118] max S , X , or or st { C 1 : ∑ n = 1 N ∑ m = 1 M b m [ n ] R ¯ l b [ n ] t ≥ or C 2 : ‖ q u [ n + 1 ] − q u [ n ] ‖ ≤ v m a x t , ∀ u ∈ U , n ∈ N C 3 : 2 ( q u j [ n ] − q m j [ n ] ) T × ( q u [ n ] − q m [ n ] ) − ‖ q u j [ n ] − q m j [ n ] ‖ 2 ≥ ( d s ) 2 C 4 : q u [ 1 ] = q u [ N ] , ∀ u ∈ U , n ∈ N

[0119] at this time It is a convex optimization problem and can be solved by using corresponding convex optimization tools.

[0120] Step S5: Repeat steps S2, S3, and S4. As the three sub-problems are optimized alternately, the system throughput is continuously optimized iteratively. Figure 2 The algorithm flow chart shows that by setting the iteration conditions, when the system throughput after iterative optimization of the three sub-problems reaches the preset number of iterations or threshold, the iteration stops.

[0121] Step S6: After the iteration stops, return to the optimal user association strategy , UAV transmission power 、Drone flight trajectory In this way, we get the approximate optimal solution of the original optimization problem.

[0122] Figure 3 Demonstrated drone base station service cycle , Communication Energy Budget The initial and optimized trajectories of the two drones are shown in Figure 2. The initial deployment positions of the two drones are both (1000, 1000) and they fly along circular trajectories within the mission cycle constraints and finally return to the initial positions. According to the constraints, the two drones continuously adjust their trajectories, transmission power, and association strategies with ground users within the mission cycle to maximize the minimum throughput of the system.

[0123] When a drone is moving, the channel conditions between it and ground users change as its position in the air changes. Therefore, the user association strategy of the drone needs to be optimized in each time slot to improve the communication service performance. Figure 4 and Figure 5 The user association of two drones in the entire service cycle (30 time slots) is shown. b u , k [ n ] = 1 It means that the drone and the ground user establish a connection in the corresponding time slot, and the ground user can obtain the drone's service at this time. In addition, it can be seen that drone 1 mainly provides services to GU-3, GU-4, and GU-5 ground users, especially maintaining long-term connections with GU-4 and GU-5. Drone 2 mainly maintains long-term connections with GU-1, GU-2, and GU-3, especially with GU-1 and GU-2. This is determined by the relative positions of the two drones and the ground users. The two drones cooperate to provide communication services for 5 ground users.

[0124] Figure 6The optimal transmission power strategy of the two drones in each time slot is shown. As the drone's air position changes, the channel conditions between the drone and the ground user also change. Figure 4 and Figure 5 After the association strategy determines the connection with the ground user, it is necessary to adjust its own transmission power according to the channel conditions in the current time slot, and the transmission power of each drone is strictly constrained by the communication energy budget. Drone 1 mainly serves GU-3, GU-4, and GU-5, and Drone 2 mainly serves GU-1, GU-2, and GU-3. The differences in serving ground users and different channel conditions make the transmission power of the two drones show different characteristics.

[0125] Figure 7 The optimized trajectories of the two drones are shown. Since the drones in this embodiment are deployed at a fixed height, only the horizontal position of the drones is shown. Compared with the initial trajectory, under the strict mission cycle constraints, the drones are closer to the ground users, and drone 2 even flies above GU-2 for 3 time slots. At the same time, a safe flight distance is also guaranteed between the two drones, demonstrating the effectiveness of the method of the present invention.

[0126] Figure 8 The system throughput of drones at different deployment altitudes is shown. It can be seen that as the deployment altitude of drones increases, the system throughput decreases. This is because as the altitude of drones increases, the path loss between drones and ground users modeled by the free space loss model increases, thus affecting the system performance. In addition, when the deployment altitude of drones is reduced from 300 meters to 200 meters, the system throughput does not improve much. This is due to the mutual influence between different association strategies of drones and channel conditions, which has certain guiding significance for the altitude deployment of drones.

[0127] Figure 9 The system throughput of the UAV under different service times is shown. It can be seen that as the UAV mission cycle increases, the system throughput does not always increase. This is due to the mutual influence between the optimal aerial position of the UAV under the corresponding mission cycle and the UAV transmission power under a strict energy budget. This has certain guiding significance for the design of UAV-assisted wireless systems.

[0128] Figure 10 The system throughput of the UAV under different algorithms changes with the number of iterations. Figure 10It can be seen that with the increase of the number of iterations, the system throughput of the five algorithms continues to increase and tends to be stable, showing good convergence. Among them, the random power algorithm converges the fastest. The method of the present invention achieves the highest system throughput due to the joint optimization of the user association, transmission power and trajectory of the drone, showing superior performance.

[0129] Figure 11 The performance of different algorithms under different communication energy budgets is demonstrated. It can be seen that with the increase of energy budget, the system throughput under the five algorithms continues to increase. In particular, the alternating optimization algorithm of the present invention significantly improves the system throughput and performance through the joint optimization of user association, UAV transmission power and trajectory, and more refined management and use of resources, reflecting the superiority of the design.

[0130] The method of the present invention can be used to precisely manage the resources of the UAV-assisted wireless emergency communication system in the air-ground integrated network, thereby having the following benefits:

[0131] 1. Improve system capacity and resource allocation efficiency: By optimizing power control, drone air service location and user association strategy, drone-assisted wireless systems can adapt to uncertain user needs and network loads, thereby improving system throughput and avoiding energy waste. This optimization is particularly suitable for high-load environments such as disaster areas, ensuring that more users get stable connections.

[0132] 2. Improve network robustness and link reliability: In the face of complex terrain and dynamically changing environments, sophisticated resource management can reduce communication blind spots and improve the stability of communication links between drones and users and other nodes by adjusting the drone's air service trajectory and communication strategy in real time.

[0133] 3. Promote industrial upgrading: Accurately manage the communication resources and energy resources of the drone-assisted communication system, making it more efficient in collaborating with near-Earth satellites, providing technical support for post-disaster recovery, remote monitoring and low-latency tasks, and accelerating the deep integration of ground-based networks, space-based networks and air-based networks, providing important support for the beautiful vision of 6G green communications.

[0134] In summary, based on the downlink communication energy budget of each drone, the present invention models the system throughput maximization problem of the multi-drone assisted wireless emergency communication system as a mathematical optimization problem under multiple constraints. The original non-convex optimization problem is decomposed into three sub-problems of user association strategy optimization between drones and ground users, drone transmission power strategy optimization and drone trajectory optimization based on the block coordinate descent idea. Then, according to the mathematical characteristics of the sub-problems, the non-convex sub-problems are transformed into convex problems that can be solved by corresponding convex optimization tools by introducing auxiliary variables and continuous convex approximation and other mathematical means. The three sub-problems are iteratively solved by alternating optimization to obtain an approximate optimal solution to the original problem.

[0135] The present invention realizes the fine management of the energy of the drone-assisted emergency communication system. By jointly optimizing the user association strategy, the drone transmission power and the drone air trajectory, the minimum throughput of the system is maximized. Under the strict limitation of the drone's downlink communication energy, the communication capacity of the drone-assisted emergency communication system is effectively improved, which has important practical significance. The present invention can effectively solve the drone trajectory and resource allocation problems in the multi-drone-assisted wireless emergency communication system, realize accurate management of communication energy, and effectively improve the capacity of the system.

Claims

1. A multi-UAV assisted wireless emergency communication optimization method, characterized in that: The method is used to optimize the throughput of a multi-UAV-assisted wireless emergency communication system, including the following steps: Step S1: construct an optimization problem of maximizing the minimum system throughput under multiple constraints in a multi-UAV assisted wireless emergency communication system, and decompose the problem into three sub-problems according to the block coordinate descent method, wherein the first sub-problem is to solve the optimal user association strategy of the UAV under a fixed UAV trajectory and transmission power strategy, the second sub-problem is to solve the optimal transmission power of the UAV under a fixed UAV trajectory and user association strategy, and the third sub-problem is to solve the optimal trajectory under a fixed UAV power transmission strategy and user association strategy; Step S2: Based on the initial trajectory and power of the drone, solve the first sub-problem and obtain the user association strategy ; Step S3: Based on user association strategy and the initial trajectory of the UAV, solve the second sub-problem and obtain the UAV power transmission strategy ; Step S4: Based on user association strategy and UAV power transmission strategy , solve the third sub-problem and get the trajectory of the drone ; Step S5: Repeat steps S2, S3, and S4 to alternately solve the three sub-problems to achieve continuous iterative optimization of the system throughput. When the preset conditions are met, the optimal user association strategy is returned. , Optimal UAV launch power strategy , Optimal UAV flight trajectory .

2. The multi-UAV assisted wireless emergency communication optimization method according to claim 1 is characterized in that: The multi-UAV assisted wireless emergency communication system is A drone is used as an aerial base station to provide Providing communication services to individual users, The drone is connected to the core network through a low-orbit satellite. In the process, the UAV moves in a circle at a fixed height, the position of the ground user is known and unchanged, and the mission time is divided into time slots, and the length of each time slot is , the position of the UAV remains approximately unchanged in each time slot.

3. The multi-UAV assisted wireless emergency communication optimization method according to claim 2 is characterized in that: In step S1, the optimization problem of maximizing the minimum system throughput under multiple constraints in the multi-UAV assisted wireless emergency communication system is constructed, and the specific method is as follows: The path between the UAV and the ground user is modeled as a free space path loss model. With ground users The channel gain between is: in, For the reference distance The channel power gain at For drones With ground users The distance between the two, when the drone is at a fixed altitude, In each time slot, the drone With ground users The distance between is expressed as: in, For drones In time slot location, For users In time slot location, is the flight altitude of the drone. According to Shannon's theorem, the drone With ground node In time slot The transmission rate within is: in, is the system bandwidth, For drones In time slot The transmission power, Indicates the reference distance The signal-to-noise ratio at is the Gaussian white noise power; In any time slot, each drone base station provides services to at most one ground user. At the same time, in any time slot, each ground user can only communicate with one drone base station. To represent the drone base station With ground users The relationship between them; when the drone base station With ground users When establishing a connection, there is ,otherwise ; During the service period of the drone base station, the system throughput is expressed as: in, is the time slot length; As an aerial communication platform, the drone base station provides wireless services to ground users. The communication energy consumption of a drone base station during the entire service cycle is: Introducing auxiliary variables represents the minimum throughput of the system, and the system throughput maximization problem based on the UAV communication energy budget is modeled as: in, , and are user association, UAV transmission power and UAV air position set respectively; C2-C4 are user association constraints; is the maximum transmission power of the UAV, is the communication energy budget of the UAV, C5 and C6 are the transmission power and information transmission energy constraints of the UAV; is the maximum flight speed of the drone, is the anti-collision distance between UAVs, C7-C9 are the trajectory constraints of the UAV base station, where C7 constrains the speed of the same UAV between adjacent time slots, C8 constrains the safety distance between different UAVs, and C9 is a circular trajectory constraint to ensure that the UAV base station returns to the initial position in the end.

4. The multi-UAV assisted wireless emergency communication optimization method according to claim 3 is characterized in that: In step S1, the first sub-problem is expressed as: The second sub-problem is expressed as: The third sub-problem is expressed as: 。 5. The multi-UAV assisted wireless emergency communication optimization method according to claim 4 is characterized in that: The specific method of step S2 is: Binary variables Relaxation is a continuous variable, then the first subproblem is expressed as: This problem is a standard linear programming problem. It is solved using the mathematical tools of linear programming and the optimization toolkit to obtain the user's associated solution. .

6. The multi-UAV assisted wireless emergency communication optimization method according to claim 4 is characterized in that: In step S3, the second sub-problem is a standard linear programming problem, which is solved using the mathematical tools of linear programming and the optimization toolkit to obtain the UAV base station power transmission strategy. .

7. The multi-UAV assisted wireless emergency communication optimization method according to claim 4 is characterized in that: The specific method of step S4 is: In the third subproblem, C1 and C8 are non-convex constraints; For the non-convex constraint C1, Lieutenant General As a whole, About A quasi-convex function, given any local point By first-order Taylor expansion and sorting, we can get The lower bound : in: Then C1 is expressed as: Put C8 in and Perform a first-order Taylor expansion at The global lower bound of : Thus, C8 is transformed into a linear constraint; At this point, the third sub-problem is expressed as: This problem is a standard convex optimization problem. The interior point method is used to solve it and the trajectory of the drone is obtained. .

Images