A UAV-assisted coverage method and system based on continuous convex approximation

Abstract

This invention proposes a UAV-assisted coverage method and system based on continuous convex approximation, comprising: Step 1: Constructing distance vectors from the UAV to each node based on the coordinates of ground nodes, and establishing a three-dimensional non-convex optimization problem in conjunction with antenna beam constraints; Step 2: Calculating node spacing and constructing equivalent nodes based on the coordinates of ground nodes, and reducing the three-dimensional non-convex optimization problem to a two-dimensional planar coordinate positioning problem based on the distance vectors from the UAV to the equivalent nodes; Step 3: Iteratively solving the two-dimensional planar coordinate positioning problem using a continuous convex approximation algorithm to obtain a two-dimensional local optimum solution; Step 4: Mapping the two-dimensional local optimum solution back to the UAV's three-dimensional coverage coordinate solution set; Step 5: Calculating the UAV antenna azimuth angle variable function and the UAV antenna downtilt angle variable function based on the UAV's three-dimensional coverage coordinate solution set and the ground node coordinates, respectively. This invention significantly improves the spatial deployment efficiency and coverage performance of UAVs, and reduces the deployment and computational overhead of communication systems.

Description

Technical Field

[0001] This invention belongs to the field of UAV communication technology, and particularly relates to a UAV-assisted coverage method and system based on continuous convex approximation. Background Technology

[0002] With the rapid development of drone technology, its application in the field of wireless communication is becoming increasingly widespread. Utilizing drones to provide aerial communication coverage has become an important means of enhancing terrestrial network capabilities, improving coverage range, and increasing flexible deployment capabilities. In the development trend of future communication networks, providing efficient and flexible wireless coverage for ground nodes using drones equipped with directional antennas has significant application prospects.

[0003] However, improving the efficiency of UAV communication coverage remains a significant challenge in practical deployments. Traditional UAV coverage solutions often neglect fairness among nodes, resulting in some nodes remaining in communication blind spots for extended periods, impacting overall service quality. In UAV scenarios with directional antennas, the coverage area is influenced by the antenna beamwidth, pitch angle, and UAV position, further complicating the trade-off between coverage efficiency and fairness. Particularly, when providing communication services to a limited number of ground nodes, optimizing the UAV's position and antenna orientation to minimize the maximum communication distance while ensuring coverage of two nodes becomes a critical factor affecting system service fairness and transmission efficiency. This problem is essentially a non-convex joint optimization problem in three-dimensional space, characterized by high solution difficulty and weak interpretability. Currently, there is a lack of effective methods with clear theoretical analysis, low computational complexity, and the ability to obtain a globally optimal solution. Therefore, a novel joint design scheme for UAV spatial position and antenna orientation is urgently needed, capable of significantly improving communication fairness and coverage efficiency while ensuring coverage requirements, reducing system complexity, and thus promoting the efficient deployment and application of UAV communication systems in real-world scenarios. Summary of the Invention

[0004] To address the problems of high complexity, low efficiency, and insufficient fairness in existing UAV communication coverage strategies, this invention proposes a UAV-assisted coverage method and system based on continuous convex approximation. The aim is to significantly improve the spatial deployment efficiency and coverage performance of UAVs while reducing the deployment and computational overhead of the communication system, while ensuring fair coverage across multiple nodes.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] A UAV-assisted coverage method based on continuous convex approximation includes the following steps: Step 1: Based on a dual-ground-node UAV communication scenario, construct the distance vector from the UAV to each node according to the coordinates of the two ground nodes, and establish a three-dimensional non-convex optimization problem in combination with antenna beam constraints.

[0007] Step 2: Calculate the node distance based on the coordinates of the two ground nodes and construct equivalent nodes. Based on the distance vector from the UAV to the equivalent nodes, reduce the dimensionality of the three-dimensional non-convex optimization problem to a two-dimensional planar coordinate positioning problem. Step 3: Iteratively solve the two-dimensional planar coordinate positioning problem using a continuous convex approximation algorithm to obtain a two-dimensional local optimum solution.

[0008] Step 4: Map the two-dimensional local optimal solution back to the UAV's three-dimensional coverage coordinate solution set;

[0009] Step 5: Calculate the azimuth angle variable function and the downtilt angle variable function of the UAV antenna based on the UAV's three-dimensional coverage coordinate solution set and the ground node coordinates.

[0010] Furthermore, the dual-ground-node UAV communication scenario includes:

[0011] Unmanned aerial vehicles (UAVs) and their accessories;

[0012] Two ground nodes: ground nodes Its coordinates Ground nodes Its coordinates ;

[0013] The drone and its accessories include the drone itself, and a signal transmitter with a fixed beamwidth attached to the drone as a coverage source.

[0014] The drone trajectory variable is Variable azimuth angle of UAV antenna UAV antenna downtilt angle variable Antenna fixed lobe .

[0015] Furthermore, the three-dimensional non-convex optimization problem in step 1 for:

[0016]

[0017] in, and From drone to ground node and ground nodes The distance function; These are constraints; Calculate the 2-norm of a vector. For drone trajectory variables Axis coordinates Fixed beam lobe for antenna The cosine function.

[0018] Furthermore, the two-dimensional planar coordinate positioning problem in step 2... for:

[0019]

[0020] in, Locating coordinate variables in a two-dimensional plane coordinate system; Let be the distance vector from the UAV to the equivalent node; ground nodes and ground nodes The node spacing.

[0021] Furthermore, the coordinate variables for the two-dimensional planar coordinate positioning problem are:

[0022]

[0023] The distance vector from the drone to the equivalent node is:

[0024] .

[0025] Furthermore, step 3 includes the following sub-steps:

[0026] Step 3.1 Positioning problem based on two-dimensional plane coordinates The objective function is defined by the distance gradient function, and the first objective function is defined based on the distance gradient function. The objective function for the next iteration of the local point convexity problem;

[0027] Step 3.2 Positioning problem based on two-dimensional plane coordinates Define the original constraint numerator function and the original constraint denominator function for the local point convex problem, respectively, and calculate the gradient function of the original constraint numerator function and the gradient function of the original constraint denominator function for the local point convex problem based on the original constraint numerator function and the original constraint denominator function for the local point convex problem, respectively.

[0028] Step 3.3: Based on the original constraint numerator function and original constraint denominator function of the local point convex problem, the gradient function of the original constraint numerator function and the gradient function of the original constraint denominator function of the local point convex problem are set as follows: Constraint functions for the next iteration of the local point convex problem;

[0029] Step 3.4 Based on the aforementioned first The objective function of the local point convexity problem in the next iteration and the... The constraint function for the local point convexity problem in the next iteration is constructed. The problem of local point convexity in the next iteration;

[0030] Step 3.5 sets the convergence condition and checks if the convergence condition is met. If not, the iteration count is updated, and the process returns to step 3.2. If yes, the two-dimensional local optimum is obtained. .

[0031] Furthermore, the first Local convexity problem in the next iteration for:

[0032]

[0033] in, For the first The objective function for the next iteration of the local point convexity problem; For the first Constraint functions for the next iteration of the local point convex problem;

[0034]

[0035]

[0036] in, For drones to the first The optimal solution to the next iteration of the local point convexity problem The distance vector, Let be the distance gradient function. For the distance gradient function at the th The optimal solution to the next iteration of the local point convexity problem gradient value, For the first The optimal solution to the local point convexity problem in the next iteration; For the first Constraint function for the next iteration of the local point convex problem To fix the beam lobe of the antenna, For coordinate variables in an equivalent problem, The original constraint numerator function for the local point convex problem In the The optimal solution to the local point convexity problem in the next iteration The value, The original constraint denominator function for the local point convex problem In the The optimal solution to the local point convexity problem in the next iteration The value, The gradient function of the original constraint numerator function for the local point convex problem In the The optimal solution to the local point convexity problem in the next iteration The value, The gradient function of the original constraint denominator function for the local point convex problem In the The optimal solution to the local point convexity problem in the next iteration The value of .

[0037] Furthermore, the UAV three-dimensional coverage coordinate solution set described in step 4 for:

[0038]

[0039] in, To provide a solution set for drone-assisted coverage locations, The rotation angle is a variable whose range of values ​​is... .

[0040] Furthermore, in step 5, the azimuth angle variable function and the downtilt angle variable function of the UAV antenna are respectively:

[0041]

[0042] in, For the azimuth angle of the UAV antenna, This is a function of the UAV antenna downtilt angle. These are the coordinates of the solution set for the drone-assisted coverage location.

[0043] On the other hand, the present invention provides a UAV-assisted coverage system based on continuous convex approximation, comprising:

[0044] The 3D non-convex optimization problem construction module is used in UAV communication scenarios with two ground nodes. It constructs the distance vector from the UAV to each node based on the coordinates of the two ground nodes, and establishes a 3D non-convex optimization problem in combination with antenna beam constraints.

[0045] The optimization problem dimensionality reduction module is used to calculate the node distance based on the coordinates of two ground nodes and construct equivalent nodes. Based on the distance vector from the UAV to the equivalent nodes, the dimensionality of the three-dimensional non-convex optimization problem is reduced to a two-dimensional planar coordinate positioning problem. The solution module is used to iteratively solve the two-dimensional planar coordinate positioning problem based on a continuous convex approximation algorithm to obtain a two-dimensional local optimum solution.

[0046] The restoration module is used to map and restore the two-dimensional local optimal solution to a three-dimensional coverage coordinate solution set of the UAV.

[0047] The variable function solver module is used to calculate the UAV antenna azimuth angle variable function and the UAV antenna downtilt angle variable function based on the UAV's three-dimensional coverage coordinate solution set and the ground node coordinates.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] This invention innovatively transforms the original non-convex joint optimization problem in three-dimensional space into an equivalent two-dimensional coverage optimization problem. Specifically, firstly, by analyzing the axisymmetric structural characteristics of the original problem, a representative two-dimensional plane for covering the target area is constructed, and the UAV deployment problem in three-dimensional space is equivalently mapped onto this two-dimensional plane. Based on this, a series of locally approximate convex problems within two-dimensional adjacent regions are constructed. Through iterative solutions, the optimal two-dimensional coverage position is gradually approximated, ensuring that the algorithm theoretically possesses convergence and a controllable error bound. Furthermore, this invention proposes an efficient three-dimensional position restoration mechanism. Through bijective mapping, the optimal solution on the two-dimensional plane is restored to the UAV position in three-dimensional space. Based on the geometric coverage principle, the corresponding antenna elevation and azimuth angles are accurately calculated to achieve optimal directional coverage of the target ground node. The entire algorithm framework possesses a good mathematical structure and physical interpretation capability, significantly reducing the dimensionality and complexity of the solution space while preserving the original characteristics of the problem. This invention not only achieves a theoretical breakthrough in methodology, moving from three-dimensional non-convex optimization to two-dimensional convex approximation solutions, but also possesses good feasibility and promotional value in engineering applications, providing a solid theoretical foundation and practical path for the efficient deployment of UAV-assisted communication systems. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0051] Figure 1 This is a schematic diagram illustrating the problem relationships established in this invention;

[0052] Figure 2 This is a map showing the boundary of the feasible location domain and the distribution of the optimal location set for the UAV in this embodiment of the invention.

[0053] Figure 3 This is a diagram showing the distribution of antenna azimuth and downtilt angles at the optimal position of the UAV in an embodiment of the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0055] Example 1

[0056] The invention will now be further described with reference to the accompanying drawings.

[0057] like Figure 1 The following is a flowchart of the method of the present invention. The implementation process includes the following steps: Step 1: Based on the dual-ground node UAV communication scenario, construct the distance vector from the UAV to each node according to the coordinates of the two ground nodes, and establish a three-dimensional non-convex optimization problem in combination with antenna beam constraints.

[0058] In this embodiment, the dual-ground-node UAV communication scenario includes:

[0059] Unmanned aerial vehicles and their accessories, and two ground nodes;

[0060] The coordinates of the ground nodes are known;

[0061] The drone and its accessories include a single drone and a signal transmitter with a fixed beam lobe attached to the drone as a coverage source.

[0062] Introducing ground nodes Its coordinates Ground nodes Its coordinates Drone trajectory variables Variable azimuth angle of UAV antenna UAV antenna downtilt angle variable Antenna fixed lobe .

[0063] According to ground nodes Its coordinates Drone trajectory variables Building drones to nodes Distance vector According to ground nodes Its coordinates Drone trajectory variables Building drones to nodes Distance vector Drone to Node Distance vector Drones to Nodes Distance vector Antenna fixed lobe Construct constraint signal angle constraints. Based on the drone's arrival at the node... Distance vector Drones to Nodes Distance vector Establishing original UAV-assisted communication coverage problem based on signal angle constraints .

[0064] Preferably, step 1, based on ground nodes... Its coordinates Drone trajectory variables Building drones to nodes Distance vector for:

[0065]

[0066] Step 1, based on ground nodes Its coordinates Drone trajectory variables Building drones to nodes Distance vector for:

[0067]

[0068] Based on drone to node Distance vector Drones to Nodes Distance vector Antenna fixed lobe The constraint signal angle constraint is constructed as follows:

[0069]

[0070] in, Calculate the 2-norm of a vector.

[0071] Establishing a three-dimensional nonconvex optimization problem for:

[0072]

[0073] Step 2: Calculate the node spacing based on the coordinates of the two ground nodes and construct equivalent nodes. Based on the distance vector from the UAV to the equivalent nodes, reduce the three-dimensional non-convex optimization problem to a two-dimensional planar coordinate positioning problem.

[0074] According to ground nodes Its coordinates Ground nodes Its coordinates Calculate node spacing for:

[0075]

[0076] Step 2: Set the coordinate variables for the equivalent problem for:

[0077]

[0078] Step 2: Set ground nodes With ground nodes Equivalent nodes and for:

[0079]

[0080] Based on the coordinate variables of the equivalent problem Node spacing Calculate the distance vector from the UAV to the equivalent node in the equivalence problem. for:

[0081]

[0082] in, Let be the distance vector from the drone to the equivalent node. The node spacing, For coordinate variables in equivalent problems.

[0083] Based on the three-dimensional nonconvex optimization problem Constructing a two-dimensional plane coordinate positioning problem for:

[0084]

[0085] in, Locating coordinate variables in a two-dimensional plane coordinate system; Let be the distance vector from the UAV to the equivalent node; ground nodes and ground nodes The node spacing.

[0086] Step 3: Iteratively solve the two-dimensional planar coordinate positioning problem using the continuous convex approximation algorithm to obtain the two-dimensional local optimum solution;

[0087] Step 3 includes the following sub-steps:

[0088] Step 3.1 Positioning problem based on two-dimensional plane coordinates The objective function is defined by the distance gradient function, and the first objective function is defined based on the distance gradient function. The objective function for the next iteration of the local point convexity problem;

[0089] Positioning problem based on two-dimensional plane coordinates Define the distance gradient function for:

[0090]

[0091] The distance gradient function Definition of the first The objective function of the next iteration local point convexity problem for:

[0092]

[0093] in, Let be the distance vector from the drone to the equivalent node. For drones to the first The optimal solution to the next iteration of the local point convexity problem The distance vector, For the distance gradient function at the th The optimal solution to the next iteration of the local point convexity problem gradient value, For the first The optimal solution to the local point convexity problem in the next iteration.

[0094] Step 3.2 Positioning problem based on two-dimensional plane coordinates Define the original constraint numerator function and the original constraint denominator function for the local point convex problem, respectively, and calculate the gradient function of the original constraint numerator function and the gradient function of the original constraint denominator function for the local point convex problem based on the original constraint numerator function and the original constraint denominator function for the local point convex problem, respectively.

[0095] Positioning problem based on two-dimensional plane coordinates Define the original constraint numerator function for the local point convex problem. for:

[0096]

[0097] Positioning problem based on two-dimensional plane coordinates Define the original constraint denominator function for the local point convex problem. for:

[0098]

[0099] The gradient function of the original constraint numerator function of the local point convexity problem is defined based on the original constraint numerator function of the local point convexity problem. for:

[0100]

[0101] The gradient function of the denominator function of the original constraint in the iterative local convexity problem is defined based on the original constraint denominator function. for:

[0102]

[0103] Step 3.3 Based on the original constraint numerator function of the local point convex problem The original constraint denominator function of the local point convex problem The original constraint numerator function and gradient function of the local point convex problem The gradient function of the original constraint denominator function and the local point convexity problem Setting the first Constraint function for the next iteration of the local point convexity problem for:

[0104]

[0105] in, For the first Constraint function for the next iteration of the local point convex problem The original constraint numerator function for the local point convex problem In the The optimal solution to the local point convexity problem in the next iteration The value, The original constraint denominator function for the local point convex problem In the The optimal solution to the local point convexity problem in the next iteration The value, The gradient function of the original constraint numerator function for the local point convex problem In the The optimal solution to the local point convexity problem in the next iteration The value, The gradient function of the original constraint denominator function for the local point convex problem In the The optimal solution to the local point convexity problem in the next iteration The value, For the first The optimal solution to the local point convexity problem in the next iteration.

[0106] Step 3.4 Based on the aforementioned first The objective function of the local point convexity problem in the next iteration and the... The constraint function for the local point convexity problem in the next iteration is constructed. The problem of local point convexity in the next iteration;

[0107] No. Local convexity problem in the next iteration for:

[0108]

[0109] in, For the first The objective function of the next iteration local point convexity problem For coordinate variables in an equivalent problem, For the first The constraint function for the next iteration of the local point convex problem. This problem can be solved efficiently using various convex optimization methods.

[0110] Step 3.5 sets the convergence condition and checks if the convergence condition is met. If not, the iteration count is updated, and the process returns to step 3.2. If yes, the two-dimensional local optimum is obtained. .

[0111] The convergence condition is defined as:

[0112]

[0113] in, For the first The optimal solution to the local point convexity problem in the next iteration. For the first The optimal solution to the local point convexity problem in the next iteration. The parameter is a fixed minimum value.

[0114] Step 4: Map the two-dimensional local optimal solution back to the UAV's three-dimensional coverage coordinate solution set, such as... Figure 2 As shown;

[0115] The local optimal solution obtained in step 3.4 Restored to drone-assisted coverage location set for:

[0116]

[0117] in, To provide a solution set for drone-assisted coverage locations, The rotation angle is a variable whose range of values ​​is... , This is a locally optimal solution. ground nodes coordinate, ground nodes coordinate.

[0118] Step 5: Calculate the UAV antenna azimuth angle variable function and the UAV antenna downtilt angle variable function based on the UAV's 3D coverage coordinate solution set and ground node coordinates, respectively. The results are as follows: Figure 3 As shown.

[0119] According to ground nodes Its coordinates Ground nodes Its coordinates UAV-assisted coverage location set Calculate the azimuth angle variable function of UAV antenna for:

[0120]

[0121] in, To provide a solution set for drone-assisted coverage locations, ground nodes coordinate, ground nodes coordinate.

[0122] According to ground nodes Its coordinates Ground nodes Its coordinates UAV-assisted coverage location set Calculate the downtilt angle variable function of UAV antenna for:

[0123]

[0124] in, To provide a solution set for drone-assisted coverage locations, ground nodes coordinate, ground nodes coordinate.

[0125] Example 2

[0126] This embodiment provides a UAV-assisted coverage system based on continuous convex approximation, including:

[0127] The 3D non-convex optimization problem construction module is used in UAV communication scenarios with two ground nodes. It constructs the distance vector from the UAV to each node based on the coordinates of the two ground nodes, and establishes a 3D non-convex optimization problem in combination with antenna beam constraints.

[0128] The optimization problem dimensionality reduction module is used to calculate the node distance based on the coordinates of two ground nodes and construct equivalent nodes. Based on the distance vector from the UAV to the equivalent nodes, the dimensionality of the three-dimensional non-convex optimization problem is reduced to a two-dimensional planar coordinate positioning problem. The solution module is used to iteratively solve the two-dimensional planar coordinate positioning problem based on a continuous convex approximation algorithm to obtain a two-dimensional local optimum solution.

[0129] The restoration module is used to map and restore the two-dimensional local optimal solution to a three-dimensional coverage coordinate solution set of the UAV.

[0130] The variable function solver module is used to calculate the UAV antenna azimuth angle variable function and the UAV antenna downtilt angle variable function based on the UAV's three-dimensional coverage coordinate solution set and the ground node coordinates.

[0131] It should be understood that any parts not described in detail in this specification belong to the prior art.

[0132] It should be understood that the above description of the preferred embodiments is quite detailed, but this should not be construed as limiting the scope of protection of this invention. It is neither necessary nor possible to exhaustively describe all possible implementations. Those skilled in the art, guided by this invention, can make substitutions or modifications without departing from the scope of the claims, all of which fall within the scope of protection of this invention. The scope of protection of this invention should be determined by the appended claims.

Claims

1. A UAV-assisted coverage method based on successive convex approximation, characterized in that, Comprise the following steps: step 1: based on the communication scene of two ground nodes unmanned aerial vehicle, according to the coordinates of two ground nodes unmanned aerial vehicle to each node distance vector, combined with the establishment of three-dimensional non-convex optimization problem of antenna beam constraint, three-dimensional non-convex optimization problem Is: wherein, and are distance functions from the UAV to the ground node and the ground node respectively; are constraint conditions; is a two-norm calculation of a vector, is a UAV trajectory variable axis coordinate, is a cosine function of an antenna fixed lobe . Step 2: Calculate the inter-node distance based on the coordinates of the two ground nodes and construct equivalent nodes, and reduce the three-dimensional non-convex optimization problem to a two-dimensional plane coordinate positioning problem based on the distance vector of the unmanned aerial vehicle to the equivalent nodes; two-dimensional plane coordinate positioning problem is: wherein, is a two-dimensional plane coordinate positioning problem coordinate variable; is a UAV to equivalent node distance vector; is a ground node and ground node node distance; step 3: iteratively solving the two-dimensional plane coordinate positioning problem based on a continuous convex approximation algorithm to obtain a two-dimensional local optimal solution; Step 4: mapping the two-dimensional local optimal solution to a set of three-dimensional coverage coordinate solutions of the UAV; Step 5: calculating a variable function of the azimuth angle of the antenna of the UAV and a variable function of the downtilt angle of the antenna of the UAV respectively according to the set of three-dimensional coverage coordinate solutions of the UAV and the coordinates of the ground nodes.

2. The UAV-assisted coverage method based on successive convex approximation of claim 1, wherein, The UAV communication scenario based on two ground nodes comprises: The UAV and its accessories; Two ground nodes: ground node with its coordinates , ground node with its coordinates ; Among the UAV and its accessories, the UAV and a signal transmitter with a fixed lobe hung under the UAV as a coverage source are included. UAV trajectory variable , UAV antenna azimuth variable , UAV antenna downtilt variable , antenna fixed lobe .

3. The UAV-assisted coverage method based on successive convex approximation of claim 1, wherein, The coordinate variables of the two-dimensional plane coordinate positioning problem are: The distance vector of the UAV to the equivalent node is: 。 4. The UAV-assisted coverage method based on successive convex approximation of claim 3, wherein, The step 3 comprises the following sub-steps: Step 3.1 Setting a distance gradient function based on a target function of a two-dimensional planar coordinate positioning problem Step 3.2 Setting a target function of a first iteration local point convex problem based on the distance gradient function Step 3.3 Setting a target function of a second iteration local point convex problem based on the target function of the first iteration local point convex problem Step 3.2 Locating according to two-dimensional plane coordinate problem The original constraint numerator function and the original constraint denominator function of the local point convex problem are respectively set, and the original constraint numerator function gradient function and the original constraint denominator function gradient function of the local point convex problem are calculated based on the original constraint numerator function of the local point convex problem and the original constraint denominator function of the local point convex problem respectively. Step 3.3 sets the first iteration local point convex problem constraint function according to the local point convex problem original constraint numerator function and the local point convex problem original constraint denominator function, the local point convex problem original constraint numerator function gradient function and the local point convex problem original constraint denominator function gradient function. function; Step 3.4 constructing the first iteration local point convex problem objective function based on the first iteration local point convex problem constraint function and the first iteration local point convex problem objective function Step 3.5 constructing the first iteration local point convex problem constraint function based on the first iteration local point convex problem objective function and the second iteration local point convex problem objective function Step 3.5 constructing the first iteration local point convex problem constraint function based on the first iteration local point convex problem objective function and the second iteration local point convex problem objective function Step 3.5 constructing the first iteration local point convex problem constraint Step 3.5 sets a convergence condition, judges whether the convergence condition is established, if not, the iteration number is updated, and returns to step 3.2; if yes, a two-dimensional local optimal solution is obtained .

5. The UAV-assisted coverage method based on successive convex approximation of claim 4, wherein, The first Subiteration local point convex problem is: wherein, is the objective function of the th iteration of the local point-convex problem; is the constraint function of the th iteration of the local point-convex problem; wherein, is the distance vector from the drone to the optimal solution of the local point-convex problem at the optimal solution of the local point-convex problem at the is the distance gradient function, is the gradient value of the distance gradient function at the optimal solution of the local point-convex problem at the optimal solution of the local point-convex problem at the is the optimal solution of the local point-convex problem at the optimal solution of the local point-convex problem at the is the constraint function of the local point-convex problem at the optimal solution of the local point-convex problem at the is the antenna fixed lobe, is the equivalent problem coordinate variable, is the original constraint numerator function of the local point-convex problem at the optimal solution of the local point-convex problem at the optimal solution of the local point-convex problem at the optimal solution of the local point-convex problem at the is the original constraint denominator function of the local point-convex problem at the optimal solution of the local point-convex problem at the optimal solution of the local point-convex problem at the optimal solution of the local point-convex problem at the is the original constraint numerator function gradient function of the local point-convex problem at the optimal solution of the local point-convex problem at the optimal solution of the local point-convex problem at the optimal solution of the local point-convex problem at the is the original constraint denominator function gradient function of the local point-convex problem at the optimal solution of the local point-convex problem at the optimal solution of the local point-convex problem at the optimal solution of the local point-convex problem at the 6. The UAV-assisted coverage method based on successive convex approximation of claim 5, wherein, The three-dimensional coverage coordinate solution of the unmanned aerial vehicle described in step 4 Is: wherein, is a solution of the UAV assisted coverage location, is a rotation angle variable, which takes a value range of .

7. The UAV-assisted coverage method based on successive convex approximation of claim 6, wherein, The variable functions of the azimuth angle of the antenna of the UAV and the downtilt angle of the antenna of the UAV in the step 5 are respectively: wherein, is a variable function of the azimuth angle of the UAV antenna, is a variable function of the downtilt angle of the UAV antenna; are the coordinates of the auxiliary coverage position solution set of the UAV, respectively.

8. A UAV-assisted coverage system based on successive convex approximations, characterized in that, Comprise: The three-dimensional non-convex optimization problem construction module is used to construct the distance vector of the UAV to each node based on the UAV communication scenario based on two ground nodes according to the coordinates of the two ground nodes, and to establish a three-dimensional non-convex optimization problem in combination with the antenna beam constraint; The optimization problem dimension reduction module is used to calculate the node distance based on the coordinates of the two ground nodes and to construct an equivalent node, and to reduce the three-dimensional non-convex optimization problem to a two-dimensional plane coordinate positioning problem based on the distance vector of the UAV to the equivalent node; The solving module is used to obtain a two-dimensional local optimal solution by iteratively solving the two-dimensional plane coordinate positioning problem based on a continuous convex approximation algorithm; The reduction module is used to map the two-dimensional local optimal solution to a set of three-dimensional coverage coordinate solutions of the UAV; The variable function solving module is used to calculate a variable function of the azimuth angle of the antenna of the UAV and a variable function of the downtilt angle of the antenna of the UAV respectively according to the set of three-dimensional coverage coordinate solutions of the UAV and the coordinates of the ground nodes. The UAV assisted coverage system based on the continuous convex approximation is used to perform the steps in the UAV assisted coverage method based on the continuous convex approximation in any one of claims 1-7.

Images