Trajectory planning and power control method for communication-assisted UAV that can be charged during flight

Abstract

The present invention discloses a trajectory planning and power control method for an auxiliary communication UAV that can be charged during flight, which is applied to the fields of mobile communication and artificial intelligence. It aims to solve the problem that it is far from enough to rely solely on ground-based communication facilities to provide services to users when there are excessive users. The present invention uses a small UAV that can be charged during flight as an aerial access node to assist ground communication, so that the UAV can provide users with reliable communication access and forwarding services for a long time, thereby effectively supplementing the capacity and coverage capability of base stations.

Description

Technical Field

[0001] The present invention belongs to the field of mobile communication and artificial intelligence, and in particular relates to a drone-assisted communication technology. Background Art

[0002] In recent years, mobile communication technology has developed rapidly and has been widely used. However, in special circumstances such as sudden disasters, system failures, and large-scale gatherings, the ground base station capacity may be insufficient. In this case, the ground base station can only serve a part of the users in the cell where the base station is located, and the remaining users cannot obtain effective services because they exceed the base station capacity. We call these users the excess users of the cell / base station.

[0003] In the case of excessive users, it is far from enough to rely solely on ground-based communication facilities to provide services to users. In order to meet user needs, enhance wireless coverage, and increase user capacity, drones can be deployed as temporary base stations or relays to assist ground users in wireless communications. Using drones to assist ground communications has the following advantages:

[0004] (1) Line-of-sight channel: Since the channel between the UAV and the ground user is mainly a direct link without refraction or scattering, the channel communication quality between the UAV and the ground communication equipment is high and the attenuation is small, which can effectively improve the signal-to-noise ratio at the receiving end, thereby improving the communication quality of the user.

[0005] (2) High mobility: Drones are flexible and maneuverable, and their locations can be adjusted in real time as needed to achieve emergency communications.

[0006] (3) Low-cost networking: A drone swarm consisting of multiple drones can quickly build a stable temporary communication network in different application scenarios and can be redeployed multiple times. Therefore, drones can be used for low-cost temporary networking to meet different types of communication needs.

[0007] Therefore, drone-assisted communication is one of the key directions of current 5G / 6G research. However, the battery capacity of drones is very limited and it is difficult to support the completion of long-term missions. Therefore, the endurance problem has always been an important factor that restricts the ability of drone-assisted communication. In order to extend the life of drone networks, existing research usually chooses to replace batteries for drones, land or hover over charging platforms to charge, but the process of returning to the charging station, charging or replacing batteries will cause mission interruptions and consume a lot of time.

[0008] In recent years, the medium and long-distance wireless charging technology for drones has been developing and maturing. In 2017, Huawei proposed at the Mobile World Congress (MWC) that base stations could be used to charge drones, and hoped that drones would never land in the future. In 2020, Huawei obtained a patent for wireless charging of mobile phones through lasers. This technology will allow multiple users with supported devices to charge through wireless modules. This technology can be used for indoor short-distance charging from mobile phones to drones and other devices, and can also be used for outdoor medium and long-distance charging. In 2018, an article published by Tongji University proposed a distributed laser charging solution that can provide watt-level power transmission at meter-level distances for the Internet of Things (IoT) and mobile devices. In January 2020, Xiaomi launched the MI AirCharge technology for wireless charging. Mobile phones can receive millimeter-wave signals sent by charging piles in the air, realizing 5-watt medium and long-distance charging within a radius of several meters. The charging piles have 5 built-in phase interference antennas, which can perform "millisecond-level" spatial positioning of mobile phones. In October 2021, Ericsson worked with laser innovation company PowerLight Technologies to successfully activate a completely exhausted 5G millimeter wave base station in Seattle through laser technology, achieving the transmission of 480 watts of energy within a distance of 300 meters. As wireless charging technology develops rapidly, drone equipment is also constantly leaning towards lightweight. DJI's smallest model currently on sale, DJI's DJ1Mini2, weighs only about 245g and has a flight power of about 250mw. This provides favorable conditions for realizing the idea of ​​long-distance wireless charging of drones in flight.

[0009] The rapid development and maturity of wireless charging technology has made it possible for drones to charge while flying. As a result, drones can simultaneously perform charging and communication service tasks during flight. This breaks through the problem that in traditional technologies, drone communication services and charging behaviors are independent of each other and cannot be performed in parallel, and opens up new application scenarios and application models for drone-assisted communication technology.

[0010] Therefore, in view of the new application scenario and mode in which charging and communication services are provided in parallel during the flight of this type of drone, the flight trajectory and power optimization mechanism and algorithm are designed for the case in which charging and communication services are provided in parallel, so that the drone can provide users with reliable communication services for a long time without interruption. This is of great significance for improving the communication quality of excess users in a cell and effectively supplementing the capacity and coverage capability of base stations. Summary of the invention

[0011] In order to solve the above technical problems, the present invention proposes a method for trajectory planning and power control of an auxiliary communication UAV that can be charged during flight, and uses a small UAV that can be charged during flight as an aerial access node to assist ground communication, so as to provide communication services for excess users; and based on DQN, the trajectory of the UAV is planned and the transmission power is controlled, so as to optimize the throughput of the UAV auxiliary communication system without exhausting the energy of the UAV, interrupting the mission, and having no blind spots in coverage.

[0012] The technical solution adopted by the present invention is: a trajectory planning and power control method for an auxiliary communication UAV that can be charged during flight, and the system based on it specifically includes: a user, a GBS (Ground Base Station), a charging device deployed on the GBS, and an unmanned aerial vehicle (UAV); the GBS is used to provide communication services to users in its coverage area, and the charging device is used to provide wireless charging for the UAV;

[0013] When the number of users in the area covered by GBS is greater than the capacity limit of GBS, drones are used to provide access services for the users that exceed the capacity limit of GBS. The specific implementation process of using drones to provide access services for the users that exceed the capacity limit of GBS is as follows:

[0014] Under the constraints of maximum transmit power, space, user access and UAV remaining energy threshold, a DQN model for UAV trajectory planning and power control is established to maximize the throughput of the UAV-assisted communication system.

[0015] The drone completes the access service based on the DQN model.

[0016] Beneficial effects of the present invention: The method of the present invention uses a small drone that can be charged during flight as an aerial access node to assist ground communication and provide communication services for excess users. Based on DQN, the trajectory and transmission power of the drone are planned to optimize the throughput of the drone-assisted communication system without exhausting the drone's energy, interrupting the mission, and covering without blind spots. It has the following advantages:

[0017] (1) A trajectory planning and power control model for charging and communication services during UAV flight was constructed. This model enables wireless charging of UAVs during flight without interrupting auxiliary communication tasks, effectively improving the life of the UAV network and reducing the impact of UAV power consumption on mission execution efficiency.

[0018] (2) A joint optimization algorithm of trajectory planning and power control based on DQN is proposed, which maximizes the system throughput of the entire UAV-assisted communication system without exhausting energy, interrupting missions, and having blind spots in coverage. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a diagram of the UAV-assisted ground excess user communication scenario;

[0020] Figure 2 This is a schematic diagram of the division of the charging area of ​​the base station;

[0021] Figure 3 This is a flow chart of the joint optimization algorithm of trajectory planning and power control based on DQN. DETAILED DESCRIPTION

[0022] In order to facilitate those skilled in the art to understand the technical content of the present invention, the following technical terms are first explained:

[0023] 1. Excessive users

[0024] In the event of a sudden disaster, system failure, or large-scale gathering, the ground base station may be insufficient in capacity. At this time, the ground base station can only serve a portion of the users in the cell where the base station is located, and the remaining users cannot receive effective services because they exceed the base station capacity. These users are called excess users of the cell / base station.

[0025] When there are excessive users, it is far from enough to rely solely on ground communication facilities to provide services to users. In order to meet user needs, enhance wireless coverage, and increase user capacity, drones can be deployed as temporary base stations or relays to assist ground users in wireless communication.

[0026] The present invention designs a method of using a small drone that can be charged during flight as an aerial access node to assist ground communication, providing communication services for excess users. Based on DQN, the trajectory and transmission power of the drone are planned, and the throughput of the drone-assisted communication system is optimized under the condition that the drone energy is not exhausted, the mission is not interrupted, and there is no blind spot in the coverage. Those skilled in the art should know that DQN is a Q-learning algorithm based on deep learning.

[0027] The following is a detailed description of the technical solution:

[0028] 1. Scenario description

[0029] like Figure 1 As shown in Figure 1, in a certain cell, there is a ground base station (GBS) G and n mobile users. The position of GBS G is regarded as the origin of the area, that is, the horizontal coordinate x G =0, ordinate yG =0, height coordinate h G =0, abbreviated as (0,0,0). GBS can provide communication services for users in the cell, but due to the capacity limitation of GBS, it can only provide access services for n0 users. When n≤n0, all users can obtain effective communication services from GBS; when n≥n0, the remaining k=n-n0 users need to use the access services provided by drones. These mobile users who need to access through drones are denoted as u j (j∈{1,2,…,k}), the set composed of them is denoted as U={u1,u2,…,u k}. Assume that there are m drones (m<<k) that can provide access services, denoted as v i (i∈{1,2,…,m}), the set it consists of is denoted as V={v1,v2,…,v m}.

[0030] In order to ensure the real-time and continuity of drone services and effectively improve the endurance of drones, it is assumed that drones can be wirelessly charged during flight. The charging equipment is deployed on the GBS and can provide energy to the drone by electromagnetic radiation. The entire charging process does not require the drone to dock on the GBS. GBS can implement aerial wireless charging for one or more drones at the same time according to the needs of the drone network. It can meet the charging needs of up to M drones at the same time. For simplicity, it is assumed that M is not less than the number of drones (i.e., M≥m).

[0031] The number of users served by a drone during a mission is determined by its location and coverage radius. To achieve better coverage, each drone needs to adjust its own position and change its transmission power to adjust its coverage radius. Assuming that each drone can provide services to a maximum of K users, the i-th drone v i The set of users served is denoted as U i , then U 1 ∪U 2 ∪…∪U m =U. Use k i Indicates drone v i The number of users served, k1+k2+…+k m = k. Considering the service efficiency, it is assumed that each user can only access the service of one drone at a time, even if it is within the coverage radius of multiple drones at the same time. That is, For the convenience of calculation, this embodiment divides time into sufficiently small time slots with length t0 as the unit. The position of the drone in the same time slot is regarded as a constant value, and the time slot number is represented by t(t∈{1,2,...,T}). iIn a given time slot t, the number of users served is k i is also a fixed value. The effective service time obtained by each served user is

[0032] 2. Joint optimization model of trajectory planning and power control for parallel charging and communication services during flight

[0033] (1) Communication model between drones and ground users

[0034] The transmission link between the UAV and the ground user can be regarded as an A2G (air-to-ground) channel, which is represented by a probabilistic line-of-sight model, that is, the path loss of the channel is a probability-weighted combination of line-of-sight (LoS) and non-line-of-sight (NLoS) links. This channel model takes into account the influence of possible uncertain factors such as terrain features and buildings and vegetation. Therefore, the channel between the UAV and the ground user may be a LoS link or a NLoS link, where the probability P of the UAV and the user being connected via a line-of-sight link is . LoS (t) is:

[0035]

[0036] in, and ψ are two constants related to the environment; is the distance between the UAV and the user in time slot t, x v (t) and x u (t) represents the horizontal coordinates of the location of the drone and the user at time slot t, y v (t) and y u (t) represents the ordinate of the location of the drone and the user at time slot t, and h is the flight altitude of the drone. At the same time, the probability P of the drone and the user being connected via a non-line-of-sight link can be obtained. NLoS (t) = 1-P LoS (t).

[0037] When data is communicated between the drone base station and the ground user, the path loss of line-of-sight transmission and non-line-of-sight transmission is given by the following formula:

[0038]

[0039] in, x v (t) and x u (t) represents the horizontal coordinates of the location of the drone and the user at time slot t, y v (t) and y u(t) represents the ordinate of the location of the drone and the user at time slot t, h represents the height of the drone, and f is the carrier frequency. The average path loss of the A2G channel at time slot t is for:

[0040]

[0041] Therefore, the ground user u j Through drone v i The signal-to-noise ratio received when access is achieved is:

[0042]

[0043] where p vi (t) is the time slot UAV v i The transmission power of user u in time slot t j In drone v i Within the coverage radius of the drone v i For user u j Provide access service, the average path loss of A2G channel is expressed as σ 2 is the noise power. Then user u j The data rate of communication is:

[0044]

[0045] Among them, B represents the drone v i With user u j The communication bandwidth during communication. The total throughput when m drones provide access services to k users in t time slots is:

[0046]

[0047] (2) UAV wireless charging model The present invention deploys charging equipment on the base station and uses electromagnetic radiation to provide wireless charging services for UAVs. At the same time, the base station can wirelessly charge multiple UAVs within the charging range. During the mission execution, in order to achieve better coverage, each UAV needs to constantly adjust its position, so the distance from the base station will be in constant dynamic change, and the charging efficiency it receives will also change. When the UAV v i The distance to GBS is When i In the dth charging area of ​​GBS, Since GBS is located at the origin of the region with coordinates (0,0,0), the distance between the drone and GBS is and are the horizontal and vertical coordinates of the position of the drone in time slot t, h is the flight altitude of the drone (the flight altitude of the drone remains unchanged during the mission execution, so h does not change with time), and d0 is the distance unit for dividing the charging area. When the drone is outside the Dth charging area, that is, d>D, the charging power that the drone can receive is 0. For the convenience of calculation, Figure 2 As shown in the figure, different spheres are obtained by dividing the spheres by radius. When the drones are on the same sphere, it can be considered that the charging power received by the drones from the base station is the same, for example Figure 2 The trajectory points a1 and a6 are both on the outermost sphere, so the charging power received by the drone at these two trajectory points is the same. Similarly, the charging power p received by the drone at trajectory points a2 and a5 is charge is the same, the charging power received by the base station at a3 and a4 is the same, assuming r d is the inner radius of the dth charging area. The charging power received by any position in the dth charging area is approximately p(G,r d ), whose value can be obtained by the following formula:

[0048]

[0049] Among them, p0 is the transmission power of the base station wireless charging, G s is the source antenna gain, G r Refers to the gain of the receiving antenna, L p refers to the polarization loss, λ is the wavelength, η is the efficiency of the rectifier, and β is the adjustment parameter of the Freese free-space equation.

[0050] Then the energy received by the drone in the dth charging area can be approximately expressed as r d The product of the received power at and the flight time in the loop is:

[0051] △E≈p(G,r d )×△T d (8)

[0052] Where △T d is the flight time of the UAV in the dth charging zone.

[0053] During the flight, the UAV needs to provide access services for excessive users on the ground. During this period, communication circuits, signal transmission, and processing energy consumption and other communication-related functions will consume a certain amount of energy. In this embodiment, this part of the communication power consumption, that is, the energy consumed per unit time for transmitting data when the UAV provides access services to users, is defined as e com =p vi(t), which is the transmission power of the drone. At the same time, the drone also consumes a lot of energy to maintain the altitude and normal flight of the drone. This part of energy consumption is determined by its flight state (including speed and acceleration). However, in the present invention, except for the hovering state, the flight speed of the drone remains basically unchanged. Therefore, the flight and hovering power consumption of the drone can be expressed as constants e respectively. fli and e hov That is, the energy consumed per unit time when the drone is flying and hovering. The remaining energy of the drone in time slot t can be expressed as:

[0054]

[0055] in, For drones i The initial energy of the drone v i The distance from the ground base station is When i In the dth charging area of ​​the ground base station, Since the ground base station is located at the origin of the region with coordinates (0,0,0), the distance between the drone and the ground base station is and are the horizontal and vertical coordinates of the position of the drone in time slot t, h is the flight altitude of the drone (the flight altitude of the drone remains unchanged during the mission execution, so h does not change with time), and d0 is the distance unit for dividing the charging area. When the drone is outside the Dth charging area, that is, d>D, the charging power that the drone can receive is 0. d is the inner radius of the dth charging area, then p(G,r d ) represents the charging power that the drone can receive when it is in the dth charging area. △T d is the flight time of the drone in the dth charging zone. fl It represents the flight power consumption of the UAV, that is, the energy consumed per unit time when the UAV is flying, e com =p vi (t) represents the UAV communication power consumption, that is, the energy consumed per unit time when the UAV provides access services to users to transmit data, which is equal to the UAV's transmission power, t fli is the flight time of the drone, t hov is the hovering time of the UAV, t com It is the time when the drone provides access services to users.

[0056] (3) Optimization model

[0057] Under the constraints of space constraints, maximum transmission power constraints, drone remaining energy threshold and user access, in order to achieve the goal of maximizing the throughput of the drone-assisted communication system, the present invention establishes a joint optimization model of drone trajectory planning and power control. The optimization problem can be expressed as:

[0058]

[0059]

[0060]

[0061]

[0062]

[0063]

[0064]

[0065] Among them, (10a) and (10b) represent the location restrictions of the UAV. The UAV must fly within a limited area to avoid interfering with the normal communication of other communication devices outside the cellular cell, that is, the UAV v i The horizontal axis range is [x min ,x max ], drone v i The vertical coordinate range is [y min ,y max ]. (10c) indicates that the positions of any two UAVs in any time slot cannot be the same to ensure that there will be no collision between UAVs, that is, Indicates drone v i location, Indicates drone v j (10d) is the constraint on the transmission power of the UAV. The transmission power of each UAV shall not exceed the power limit p max At the same time, since the UAV will continuously consume energy during the flight and perform tasks, it will not be able to work when the remaining energy is too low, which will greatly reduce the efficiency of task execution. Therefore, the present invention sets a remaining energy threshold E for the UAV. dead , (10e) ensures that the drone will not fail due to energy exhaustion. (10f) indicates that any user can only receive access services from one drone at a time.

[0066] In this scenario, the user is mobile, and the drone is also adjusting its position in real time to achieve the goal of optimizing the throughput of the auxiliary communication system. The drone and the user have two-way mobility. At the same time, while optimizing the trajectory in real time, the drone is also adjusting its transmission power to optimize the coverage performance. The trajectory optimization must take into account the real-time wireless charging of the drone to balance its energy consumption and replenishment. Considering that existing studies have proved that this problem is an NP-hard problem under fewer constraints, the optimization model proposed in the present invention also considers trajectory planning, power adjustment and wireless charging at the same time, and it must be an NP-Hard problem. Not only that, the scenario faced by the model is a dynamically changing scenario, and it is impossible to directly obtain an optimal solution or a suboptimal solution, which poses a challenge to the traditional convex optimization algorithm. Therefore, the present invention adopts DQN to jointly optimize the trajectory and transmission power of the drone, and maximizes the throughput of the drone-assisted communication system without exhausting the drone's energy, interrupting the mission, and covering without blind spots.

[0067] 3. DQN-based joint optimization algorithm for trajectory planning and power control

[0068] The present invention needs to establish a DQN model for the drone, set the state space, action space and reward and punishment function of the drone. Let the state space be S, the action space be A, the reward value be R, and the triple<s,a,R> Determines the specific UAV agent model.

[0069] (1) State space

[0070] Since DQN needs to calculate the Q value of the action based on the input state information when training the model, and considering the requirements in the optimization model, the state space of the present invention needs to include the position information of the user and the drone at time slot t and Transmitting power of drone Signal-to-noise ratio when drones provide access services to users And the real-time remaining battery information of the drone That is, the set of UAV agent state spaces can be expressed as:

[0071]

[0072] (2) Action Space

[0073] Since a continuous action space will lead to an infinite action space of DQN, which increases the complexity of action selection, the action space is discretized in the present invention. The discretized action space contains two subspaces: a UAV flight action subset and a transmission power adjustment subset.

[0074] The next flight direction of the drone is divided into five options: left, right, forward, backward and hovering, which are represented by Lf(Left), Ri(Right), Fw(Forward), Bw(Backward) and Ha(Halt). When the drone flies to the boundary of the preset area, the action is considered invalid and the hovering action is performed by default. The flight action subset of the drone agent can be represented as A1={Lf,Ri,Fw,Bw,Ha}.

[0075] At the same time, the transmission power of the UAV is preset to N fixed gears for the UAV agent to choose, that is, the transmission power adjustment subset can be expressed as A2 = {p0, p1, ..., p N}.

[0076] (3) Reward and Penalty Function

[0077] Each agent receives its own reward through interaction with the environment, which depends on the current state, current action and next state (s, a, s') of each agent in each time slot environment. In the proposed drone-assisted ground user communication scenario with in-flight wireless charging, the reward and penalty function consists of three parts: throughput reward, user coverage blind spot penalty and penalty for low remaining energy. The throughput reward is expressed as In order to maximize the total throughput of the auxiliary communication system, the reward of each UAV is determined by the sum of the system rates rather than its own data rate. Under this setting, each UAV will take the action that is most conducive to improving the overall performance of the system based on the current state information to ensure the goal of maximizing the total throughput, that is, Set to:

[0078]

[0079] User override blind spot penalty with -R miss In each iteration, the ground user who failed to access will send a request to the ground base station, indicating that he failed to successfully use the drone communication. At this time, the base station sends the user's non-access information to all drones, and each drone receives the same penalty.

[0080]

[0081] Among them, c miss is a constant, and its value can be determined empirically during simulation; then the total user coverage blind spot penalty can be expressed as:

[0082]

[0083] The other part of the reward function, the low energy penalty, is expressed as If there is no energy penalty, when the drone energy is too low, in order to achieve a higher throughput, the drone will still choose to charge close to the user rather than close to the base station. Although this will increase the total throughput of the system in a certain time slot, it may eventually cause the drone to run out of energy and stop working. Therefore, the present invention designs an energy penalty as follows:

[0084]

[0085] Among them, c1, c2, c3 are constants, Denotes the time slot UAV v i The remaining energy, E low1 and E low2 are the two low energy thresholds of the drone, and their values ​​are designed to be Indicates drone v i The maximum capacity of the battery. Then v i The reward and penalty function in time slot t is:

[0086]

[0087] (4) Specific algorithm

[0088] The model of the drone consists of an interaction network Q(θ), an evaluation network Q'(θ'), an experience pool, and a gradient update function. The interaction network is responsible for interacting with the environment in real time. At time slot t, the input state s i , output action a i ; When the agent completes the action, the environment will give the next state s i+1 and reward R i , the agent converts the quaternion i ,a i ,R i ,s i+1 >Store in the experience pool; the gradient update function uses the gradient descent method to update the interaction network gradient through the quadruple obtained from the experience pool and the output of the evaluation network. The formula of the gradient descent method is After a certain number of times, the evaluation network obtains the neural network parameters from the interactive network and updates some of its own neural network parameters to ensure the accuracy of the evaluation network gradient update and the convergence of the algorithm. This cycle is repeated to finally complete the training of the deep reinforcement learning model for the drone. The pseudo code of the algorithm is as follows:

[0089]

[0090]

[0091] ​Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the scope of protection of the present invention is not limited to such specific statements and embodiments. For those skilled in the art, the present invention may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A method for trajectory planning and power control of an auxiliary communication UAV capable of charging during flight, characterized in that: The system specifically includes: users, ground base stations, charging equipment deployed on the ground base stations, and drones; the ground base stations are used to provide communication services to users in their coverage areas, and the charging equipment is used to provide wireless charging for drones; When the number of users in the area covered by the ground base station is greater than the capacity limit of the ground base station, drones are used to provide access services for the users who exceed the capacity limit of the ground base station. The specific implementation process of using drones to provide access services for the users who exceed the capacity limit of the ground base station is as follows: Under the constraints of maximum transmit power, space, user access and UAV remaining energy threshold, a DQN model for UAV trajectory planning and power control is established to maximize the throughput of the UAV-assisted communication system. The drone completes the access service based on the DQN model; The specific process of the drone completing the access service based on the DQN model is as follows: A1. Establish a DQN model, and record the state space of the drone as S, the action space as A, and the reward and punishment function as R; the state space includes: user u in time slot t j Location information Time slot t drone v i Location information Drone i Transmit power Drone i For user u j Channel bandwidth B when providing access service, drone v i For user u j Signal-to-noise ratio when providing access services And drone v i Real-time remaining energy information A2: The drone interacts with the environment and enters state s at time slot t. i , output action a i ; When the agent completes the action, the environment will give the next state s i+1 and reward R i , the agent converts the quaternion i ,a i ,R i ,s i+1 >Deposit into experience pool;​ A3, determining whether all users exceeding the capacity limit of the ground base station have accessed the network through the drone, if so, executing step A4, otherwise executing step A6; A4, determine whether the remaining energy of the drone is greater than or equal to the set threshold, if so, execute step A5, otherwise execute step A7; A5, determine whether the access service is completed, if so, end; otherwise, execute step A2; A6. Report the information of the unconnected users to the base station; A7, after calculating the energy penalty of the drone, return to step A2; The energy penalty calculation formula is: Among them, c1, c2, c3 are constants, Denotes the time slot UAV v i The remaining energy E low1 and E low2 are the two low energy thresholds of the drone, Indicates drone v i The maximum capacity of the battery; The calculation formula is: in, For drones i The initial energy of d is the inner radius of the dth charging area, d∈{1,2,...,D}, D represents the number of charging areas; p(G,r d ) represents the charging power that the drone can receive when it is in the dth charging area; △T d is the flight time of the drone in the dth charging area; e com represents the UAV communication power consumption; t fli is the flight time of the drone; e fli is the flight power consumption of the UAV; t hov is the hovering time of the drone; e hov is the hovering power consumption of the drone, t com It is the time when the drone provides access services to users.

2. The trajectory planning and power control method of an auxiliary communication UAV capable of charging during flight according to claim 1 is characterized in that: The charging equipment uses electromagnetic radiation to provide wireless charging services for drones.

3. The trajectory planning and power control method of an auxiliary communication UAV capable of charging during flight according to claim 2 is characterized in that: According to the distance from the ground base station, the space around the ground base station is discretized into D charging areas. The charging power received by drones at any position in the same charging area is the same.

Images