Dynamic spectrum sharing system and method for UAV networks based on two-stage Stackelberg game

Abstract

The present invention relates to a dynamic spectrum sharing system and method based on a two-stage Stackelberg game in a drone-assisted communication network, comprising: in the first stage of the Stackelberg game, a base station maximizes its own utility function to obtain an optimal solution of price, share reward and commission incentive, broadcasts the optimal solution to a drone, and the drone substitutes the optimal solution into its own utility function; in the second stage of the Stackelberg game, the drone substitutes the optimal solution broadcasted by the base station into its own utility function and maximizes the utility function to obtain an optimal solution of spectrum resources and price, feeds back the optimal solution of spectrum resources and price to the base station, broadcasts the optimal solution of price to a user, the user substitutes the optimal solution of price into his own utility function and maximizes the utility function to obtain an optimal solution of spectrum resources, and feeds back the optimal solution of spectrum resources to the base station and the drone; and dynamically adjusts the price, spectrum resources, commission incentive and share reward until the base station, the drone and the user obtain maximum benefits.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communications, and in particular relates to a dynamic spectrum sharing system and method for an unmanned aerial vehicle network based on a two-stage Stackelberg game. Background Art

[0002] With the increase of users and social access points, the demand for spectrum resources is also increasing. The spectrum resources that operators can provide are relatively scarce. At the same time, with the development of the times, users have higher and higher requirements for service quality. They no longer simply use the amount of shared spectrum as the criterion for evaluating the pros and cons of sharing models, but use data transmission rate as an indicator to improve the quality of service (QoS) of communication networks.

[0003] One solution is to treat drones as special users for dynamic spectrum sharing in low-altitude drone scenarios. Previously, drones have been widely used in many fields such as cargo delivery, remote sensing, data relay, and connection maintenance. Drone communications have the advantages of low cost, high flexibility, and easy scheduling, which makes drones more and more valued and entrusted with important tasks in today's society. In addition, drones can provide better communication service quality and excellent coverage capabilities. In most cases, base stations cannot provide spectrum resources to users in remote areas, nor can they guarantee the quality of spectrum resources. The excellent coverage capabilities of drones just make up for this. Drones can play a role in assisting base stations to share spectrum with users. In addition, this sharing model can achieve the goal of helping base stations expand business coverage, thereby ensuring that more users obtain spectrum resources from base stations. However, drones need to have enough motivation to participate in dynamic spectrum sharing. Therefore, it is necessary to solve the problem of encouraging more drones to participate in spectrum sharing.

[0004] A relatively small number of base stations will face an increasing number of users, which also means that the base stations will bear excessive loads during peak periods. This is not only not conducive to the base stations continuing to stably perform their expected functions, but also not conducive to users obtaining efficient and high-quality services from the base stations.

[0005] So far, there have been some studies on pricing strategies and spectrum sharing, but most of them use relatively simple game methods, such as single Stackelberg game.

[0006] Whether it is commission incentives or share rewards, the purpose is to solve the selfishness of drones in spectrum sharing. In order to further improve the status of drones, strengthen drone functions, and encourage drones to serve base stations, there are three problems that need to be solved: 1. Under what circumstances can drones get the most ideal benefits? 2. What is the total amount of spectrum resources that drones need to help base stations share? 3. How to find the most ideal balance point for base stations, drones, and users in multi-drone and multi-user scenarios? Therefore, it is necessary to consider both the overall performance of the network and the fairness of individual interests, and provide a good incentive mechanism so that these selfish drones can actively participate in spectrum sharing. This is a problem that needs to be solved urgently. Summary of the invention

[0007] In order to overcome the above-mentioned deficiencies of the prior art, the present invention provides a dynamic spectrum sharing system and method for a drone network based on a two-stage Stackelberg game, so as to improve the QoS in the communication network, reduce the load of the base station, and increase the income of the drone and the user.

[0008] According to one aspect of the present invention, a method for dynamic spectrum sharing of a drone network based on a two-stage Stackelberg game is provided, wherein:

[0009] In the first stage of the Stackelberg game, the base station maximizes its own utility function and obtains the optimal solution of price, share reward, and commission incentive, and broadcasts the optimal solution to the drone, which substitutes the optimal solution into its own utility function;

[0010] In the second stage of the Stackelberg game, the drone substitutes the optimal solution broadcast by the base station into its own utility function and maximizes its own utility function to obtain the optimal solution of spectrum resources and price, feeds back the optimal solution of spectrum resources and price to the base station, broadcasts the optimal solution of price to the user, and the user substitutes the optimal solution of price into his own utility function and maximizes his own utility function to obtain the optimal solution of spectrum resources, and feeds back the optimal solution of spectrum resources to the base station and the drone;

[0011] Through an overall iterative approach, prices, spectrum resources, commission incentives, and share rewards are dynamically adjusted until base stations, drones, and users achieve maximum benefits.

[0012] The above technical solution proposes to apply a two-stage Stackelberg game method to the dynamic spectrum sharing model. The entire dynamic spectrum sharing process is divided into two stages. In the first stage, the base station and the drone play a game; in the second stage, the drone and the user play a game. In this way, the game interaction process is more specific and clear, and a more ideal result is obtained.

[0013] The above technical solution transfers part of the functions and load of the base station, and proposes to use the share reward method to encourage more drones to act as agents of the base station to negotiate the price of data transmission rate with users and assist the base station in collecting fees paid by users. In this case, the load of the base station is reduced, and the drones and users get higher benefits.

[0014] Furthermore, the commission reward is that the drone assists the base station in sharing spectrum with users, and in return, the base station gives the drone a commission incentive.

[0015] Furthermore, the share reward is that the drone acts as an agent of the base station to negotiate the price of data transmission rate with the user and assists the base station in collecting fees paid by the user, and the base station gives the drone a share reward.

[0016] As a further technical solution, prices, spectrum resources, commission incentives, and share rewards are dynamically adjusted through overall iteration, further including:

[0017] After completing a round of iterations, the optimal solution obtained in the previous round is substituted into the utility function of the base station in the next round, and a new round of two-stage Stackelberg game is carried out, and so on, until the maximum benefits of base stations, drones, and users tend to stabilize.

[0018] As a further technical solution, in the first stage Stackelberg game, the base station is the leader and the drone is the follower.

[0019] As a further technical solution, in the second stage Stackelberg game, the drone is the leader and the user is the follower.

[0020] As a further technical solution, the utility function of the base station is defined as:

[0021]

[0022] Among them, P ni represents the unit price of spectrum resources used by UAV i, A ni represents the spectrum resources leased by base station n to drone i, B ij represents the spectrum resources shared by UAV i to user j, p ij represents the unit price of data transmission rate obtained by user j from drone i, η ni represents the share reward provided by base station n to drone i, β ni represents the commission incentive obtained by drone i for helping base station n share spectrum, SINR represents the signal-to-interference-noise ratio of data transmitted by drone i to user j, and α is the logarithmic base.

[0023] As a further technical solution, the utility function of each drone is defined as:

[0024]

[0025] Among them, P ni represents the unit price of spectrum resources used by UAV i, A ni represents the spectrum resources leased by base station n to drone i, B ij represents the spectrum resources shared by UAV i to user j, p ij represents the unit price of data transmission rate obtained by user j from drone i, η ni represents the share reward provided by base station n to drone i, β ni represents the commission incentive obtained by UAV i for helping base station n share spectrum, SINR represents the signal to interference and noise ratio of data transmitted by UAV i to user j, and α and b are both logarithmic bases.

[0026] As a further technical solution, the utility function of each user is defined as:

[0027]

[0028] Among them, B ij represents the spectrum resources shared by UAV i to user j, p ij represents the unit price of data transmission rate obtained by user j from drone i, SINR represents the signal to interference and noise ratio of data transmission from drone i to user j, and α and b are both logarithmic bases.

[0029] According to one aspect of the present invention, a dynamic spectrum sharing system for a drone network based on a two-stage Stackelberg game is provided, the system comprising:

[0030] The first stage Stackelberg game module is used to maximize the base station's own utility function, obtain the optimal solution of price, share reward, and commission incentive, and broadcast the optimal solution to the drone, which substitutes the optimal solution into its own utility function;

[0031] The second stage Stackelberg game module is used to make the drone substitute the optimal solution broadcast by the base station into its own utility function and maximize its own utility function to obtain the optimal solution of spectrum resources and price, feed back the optimal solution of spectrum resources and price to the base station, broadcast the optimal solution of price to the user, and the user substitutes the optimal solution of price into his own utility function and maximizes his own utility function to obtain the optimal solution of spectrum resources, and feed back the optimal solution of spectrum resources to the base station and the drone;

[0032] The iteration module is used to dynamically adjust prices, spectrum resources, commission incentives, and share rewards through overall iteration when the maximum benefits of base stations, drones, and users have not stabilized until base stations, drones, and users obtain maximum benefits.

[0033] As a further technical solution, the iteration module is also used to, after completing a round of iteration, substitute the optimal solution obtained in the previous round into the utility function of the base station in the next round, and conduct a new round of two-stage Stackelberg game, and so on, until the maximum benefits of the base station, drone, and user tend to stabilize.

[0034] Furthermore, the interference that drone i encounters when sharing spectrum with user j is Defined as:

[0035]

[0036] As a further improvement of the present invention, the method further includes defining the data transmission rate obtained by user j from drone i as C ij =B ij log a (1+SINR).

[0037] For users, the quality of the sharing model is no longer simply evaluated by the amount of shared spectrum. Instead, the data transmission rate is used as an indicator to obtain the corresponding optimal solution, thereby improving the quality of service (QoS) of the communication network.

[0038] Furthermore, drones help base stations share spectrum with users, and in return, the base stations will give drones a commission incentive β.

[0039] Furthermore, the drone acts as an agent of the base station to negotiate the price of data transmission rate with the user and helps the base station collect the fees paid by the user. The base station will give the drone a share reward η.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] 1. The present invention proposes a dynamic spectrum sharing system and method for drone-assisted communication. In the system and method, the drone is both a spectrum consumer, renting and using the spectrum from the base station, and a spectrum provider, sharing its idle spectrum with remote users far away from the base station.

[0042] 2. The present invention proposes a two-stage Stackelberg game, which enables the base station to find the optimal pricing plan, share reward and commission incentive, enables the drone to find the optimal spectrum resources and pricing plan, and enables the user to find the optimal spectrum resources, so as to maximize the benefits of the base station, drone and user.

[0043] 3. In order to improve QoS, the spectrum sharing performance is no longer simply measured by the total amount of spectrum shared. Instead, the data transmission rate received by users from drones is used as the performance indicator. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A block diagram of dynamic spectrum sharing in a drone-assisted communication network according to an embodiment of the present invention.

[0045] Figure 2 : is the changing relationship between the maximum value of the base station utility function and the number of iterations according to an embodiment of the present invention.

[0046] Figure 3 : is the changing relationship between the maximum value of the average utility function of the drone and the number of iterations according to an embodiment of the present invention.

[0047] Figure 4 : is the changing relationship between the maximum value of the user average utility function and the number of iterations according to an embodiment of the present invention.

[0048] Figure 5 This is a macro comparison chart of the maximum value of the average utility function of game participants between the algorithm proposed according to an embodiment of the present invention and other algorithms.

[0049] Figure 6 This is a comparison chart of the maximum values ​​of utility functions of four drones according to the algorithm proposed in an embodiment of the present invention and other algorithms.

[0050] Figure 7 This is a comparison chart of the maximum values ​​of eight user utility functions of the algorithm proposed according to an embodiment of the present invention and other algorithms. DETAILED DESCRIPTION

[0051] The following will clearly and completely describe the technical solutions of various embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0052] Example 1

[0053] In order to overcome the problems of low data quality and low spectrum efficiency caused by long distance, the present invention considers using drones as agents to help base stations share spectrum with users. Drone-assisted communication scenarios such as Figure 1 As shown in the figure, there is a base station, multiple drones and multiple users, where the drone can earn profits by assisting the base station in sharing spectrum with users, negotiating the price of data transmission rate with users as the agent of the base station and helping the base station collect fees paid by users. The specific protocol process is as follows:

[0054] (1) The base station leases spectrum resources to drones;

[0055] (2) UAVs share their idle spectrum resources with users;

[0056] (3) Drones pay operators based on the spectrum resources they use;

[0057] (4) The user pays the drone based on the data transmission rate received from the drone;

[0058] (5) The drone returns the fee paid by the user to the base station;

[0059] (6) Drones assist base stations in sharing spectrum with users, so base stations give drones commission incentives;

[0060] (7) The drone acts as the agent of the base station to negotiate the price of data transmission rate with the user and assists the base station in collecting the fees paid by the user, so the base station gives the drone a share reward.

[0061] The present invention applies a two-stage Stackelberg game method to a dynamic shared spectrum model. In the first stage, the base station is the leader and the drone is the follower. The drone selects its own optimal strategy according to the strategy selected by the base station; in the second stage, the drone is the leader and the user is the follower. The user selects its own optimal strategy according to the strategy selected by the drone. In addition, the present invention uses the augmented Lagrange multiplier method to maximize the utility functions of the base station, the drone and the user.

[0062] 1. Base station utility function

[0063] The utility function of the base station is defined as follows:

[0064]

[0065] Among them, P ni represents the unit price of spectrum resources used by UAV i, A ni represents the spectrum resources leased by base station n to drone i, B ij represents the spectrum resources shared by UAV i to user j, p ij represents the unit price of data transmission rate obtained by user j from drone i, η ni represents the share reward provided by base station n to drone i, β ni represents the commission incentive obtained by drone i for helping base station n share spectrum, SINR represents the signal-to-interference-noise ratio of data transmitted by drone i to user j, and α is the logarithmic base.

[0066] The goal of the base station is to maximize utility by solving the following optimization problem.

[0067]

[0068] Among them, constraint (a) states that the spectrum price P used by drone i itself is not less than zero, and constraint (b) indicates that the share reward η and commission incentive β provided by the base station to drone i are in the range of [0, 1].

[0069] 2. Drone Utility Function

[0070] The utility function of each drone is defined as follows:

[0071]

[0072] Where b is the logarithm base.

[0073] The goal of each UAV is to maximize its utility by solving the following optimization problem.

[0074]

[0075] Constraint (a) is the total upper limit of spectrum resources provided by the base station to all drones, and constraint (b) is the spectrum resource A provided by the base station to drone i. ni It is never less than the spectrum resource B provided by drone i to user j ij , constraint (c) represents the unit price p of the data transmission rate obtained by user j ij Not less than zero.

[0076] 3. User utility function

[0077] The utility function of each user is defined as follows:

[0078]

[0079] The goal of each user is to maximize utility by solving the following optimization problem.

[0080]

[0081] 4. Augmented Lagrange multiplier method

[0082] Optimization Problems with Inequality Constraints

[0083]

[0084] Introducing auxiliary variable z i (i=1,…,m), so that (11) can be transformed into an equivalent equality constraint problem

[0085]

[0086] At this time, the augmented Lagrangian function of (8) is

[0087]

[0088] To simplify the algorithm, let

[0089] Algorithm 1: Augmented Lagrange Multiplier Method

[0090] initialization:

[0091] Initialize the initial point x0, the initial multiplier vector λ1, the initial penalty factor μ1 and its amplification factor c>1, the control error ε>0, and the constant θ∈(0,1);

[0092] renew And set it equal to 0;

[0093] If μc i (x)-λ i ≥0,

[0094] Update z i 2 =-λ i / μ+c i (x),

[0095] Otherwise i =0;

[0096] Update z i 2 =(1 / μ)max(0,μc i (x)-λ i ), i = 1, ..., m;

[0097] z i 2 =(1 / μ)max(0,μc i (x)-λ i ), i = 1, ..., m into In

[0098] Iteration:

[0099] Initialize k=1;

[0100] Loop Start (condition: )

[0101] x 0-1 As the initial point, solve the unconstrained problem

[0102] Update the optimal solution x k ;

[0103] if

[0104] Update μ k+1 =cμ k ;

[0105] Update(λ k+1 ) i =(λ k ) i -μ[c i (x)-z i 2 ], i = 1, ..., m;

[0106] z i 2 =(1 / μ)max(0,μc i (x)-λ i ), i = 1, ..., m into

[0107] (λ k+1 ) i =(λ k ) i -μ[c i (x)-z i 2 ], i = 1, ..., m, we get

[0108] (λ k+1 ) i =max[0,(λ k ) i -μc i (x k )], i = 1, ..., m;

[0109] Update k = k + 1;

[0110] The loop ends.

[0111] Algorithm 2: Two-stage Stackelberg game

[0112] initialization:

[0113] Initialize A(0), B(0), p(0);

[0114] Iteration:

[0115] Initialize k=1;

[0116] Loop Start

[0117] The base station calculates formula (2) and updates (P(k), η(k), β(k)), and then broadcasts (P(k), η(k), β(k)) to the drone;

[0118] The drone calculates formula (4) and updates (p(k), A(k)), then broadcasts (p(k)) to the user and feeds (p(k), A(k)) back to the base station;

[0119] The user calculates formula (6) and updates (B(k)), and then feeds (B(k)) back to the base station and the drone;

[0120] Update k = k + 1;

[0121] The loop ends.

[0122] Example 2

[0123] This embodiment assumes that there is 1 operator, 4 drones and 8 users in the spectrum sharing process. The algorithm proposed in the present invention is also applicable to scenarios with more drones and users. The present invention places the system model in a three-dimensional space coordinate system. The x, y, and z axes constitute a space coordinate system, and the x and y axes constitute a plane coordinate system. The base station is placed on the z axis, each drone is in the space coordinate system and z≠0, each user is in the plane coordinate system, and the drones and users are randomly placed.

[0124] In the simulation, the channel gain G = 1, the transmission power P t =5W, noise power σ 2 =10 -14 W, path loss coefficient α = 0.5, logarithmic base α and b are both set to 2, number of iterations r is set to 20, B ij The initial value is set to the identity matrix, A ni Suppose three different initial values ​​are {[12, 11, 11, 10], [10, 10, 10, 11], [11, 12, 11, 12]}.

[0125] The proposed two-stage Stackelberg game effectively and reliably finds the maximum utility of the three types of members: base stations, drones, and users. In order to compare with the simulation results of the game method proposed in the present invention, the performance is compared with the traditional Stackelberg game method. The two-stage Stackelberg game method is referred to as the "new method" and the traditional Stackelberg game method is referred to as the "old method".

[0126] Figure 2 Given A ni The line graph of the maximum value of the base station utility function as the number of iterations increases when taking three different initial values. It can be seen that the maximum value of the base station's utility function tends to stabilize after two rounds, which is about 315-320.

[0127] Figure 3Given A ni The line graph of the maximum value of the average utility function of the drone as the number of iterations increases when taking three different initial values. It can be seen that the maximum value of the average utility function of the drone tends to stabilize after two rounds, which is about 129-131.

[0128] Figure 4 Given A ni The line graph of the maximum value of the user's average utility function as the number of iterations increases when taking three different initial values. It can be seen that the maximum value of the user's average utility function tends to stabilize after two rounds, which is about 28-33.

[0129] Figures 2 to 4 It can be seen that the maximum value of the average utility function of each member in the two-stage Stackelberg game model jumps repeatedly within a certain range. The reason is that the two-stage Stackelberg game is a special dynamic game, and the Nash equilibrium point in the game is also relatively dynamic. In each round, the second game follows the end of the first game, so the maximum value of the benefit function obtained in each round will be in an interval, which can be called the dynamic Nash equilibrium interval.

[0130] Figures 2 to 4 It can be seen that no matter A ni No matter how the initial value of changes, the convergence speed of the game model can be guaranteed, and the maximum values ​​of the members are very close.

[0131] Figures 2 to 4 It can be seen that the simulation iteration trend is fast. The reason is that the gradient descent of the augmented Lagrange multiplier method is fast, so the optimal solution can be found quickly. In addition, we simulate in Python, and the iteration value calculated internally by the Python platform is higher, making it easier to obtain the maximum value.

[0132] Figure 5 A macro comparison chart of the maximum values ​​of the average utility functions of base stations, drones, and users under the new and old methods is given. It can be seen that the maximum value of the base station utility function under the new method is 20 less than that of the old method.

[0133] Figure 6 A comparison chart of the maximum values ​​of the utility functions of four drones under the new and old methods is given. It can be seen that the maximum value of the drone utility function under the new method is 5-15 larger than that under the old method.

[0134] Figure 7 A comparison chart of the maximum values ​​of eight user utility functions under the new and old methods is given. It can be seen that the maximum value of the user utility function under the new method is 5-30 larger than that of the old method.

[0135] Figures 5 to 7It can be seen that the maximum values ​​of the utility functions of each drone and user in the new method are greater than those in the old method to varying degrees, which is in line with our expectation, that is, to increase the utility function of each member as much as possible. The utility function of the base station in the new method is slightly lower than that of the old model, because in the new method, the base station transfers the pricing task to the drone and gives the drone a certain proportion of rewards, so the utility function of the base station decreases slightly, and the load of the base station is reduced accordingly, so it is a win-win situation for the base station and the drone. For the drone, because it receives rewards from the base station, the maximum value of its utility function is increased. Why does the maximum value of the user's utility function increase? The reason is that while the maximum value of the drone utility function (3) increases, its parameter variable Compared with the old method becomes smaller, so the maximum value of the user utility function (5) is increased.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A dynamic spectrum sharing method for UAV networks based on a two-stage Stackelberg game, characterized in that: The method comprises a two-stage Stackelberg game, wherein: In the first stage of the Stackelberg game, the base station maximizes its own utility function and obtains the optimal solution of price, share reward, and commission incentive, and broadcasts the optimal solution to the drone, which substitutes the optimal solution into its own utility function; The utility function of the base station is defined as: ， The utility function of each drone is defined as: ， in, Indicates drone The unit price of the spectrum resources used, Indicates that the base station n is directed to the drone Leased spectrum resources, Indicates drone The spectrum resources shared with user j, Indicates that user j is from the drone The data transfer rate unit price is obtained, Indicates that the base station n is directed to the drone The share rewards provided, Indicates drone Commission incentives from helping base station n share spectrum, Indicates drone The signal-to-interference-to-noise ratio of data transmitted to user j, where a and b both represent logarithmic bases; In the second stage of the Stackelberg game, the drone substitutes the optimal solution broadcast by the base station into its own utility function and maximizes its own utility function to obtain the optimal solution of spectrum resources and price, feeds back the optimal solution of spectrum resources and price to the base station, broadcasts the optimal solution of price to the user, and the user substitutes the optimal solution of price into his own utility function and maximizes his own utility function to obtain the optimal solution of spectrum resources, and feeds back the optimal solution of spectrum resources to the base station and the drone; The utility function of each user is defined as: ， Through an overall iterative approach, prices, spectrum resources, commission incentives, and share rewards are dynamically adjusted until base stations, drones, and users achieve maximum benefits.

2. The method for dynamic spectrum sharing of UAV network based on two-stage Stackelberg game according to claim 1 is characterized in that: Dynamically adjust prices, spectrum resources, commission incentives, and share rewards through overall iteration, further including: After completing a round of iterations, the optimal solution obtained in the previous round is substituted into the utility function of the base station in the next round, and a new round of two-stage Stackelberg game is carried out, and so on, until the maximum benefits of base stations, drones, and users tend to stabilize.

3. The method for dynamic spectrum sharing of UAV network based on two-stage Stackelberg game according to claim 1 is characterized in that: In the first stage Stackelberg game, the base station is the leader and the drone is the follower.

4. The method for dynamic spectrum sharing of UAV network based on two-stage Stackelberg game according to claim 1 is characterized in that: In the second-stage Stackelberg game, drones are leaders and users are followers.

5. A dynamic spectrum sharing system for drone networks based on a two-stage Stackelberg game, used to implement the method described in any one of claims 1 to 4, characterized in that: The system comprises: The first stage Stackelberg game module is used to maximize the base station's own utility function, obtain the optimal solution of price, share reward, and commission incentive, and broadcast the optimal solution to the drone, which substitutes the optimal solution into its own utility function; The second stage Stackelberg game module is used to make the drone substitute the optimal solution broadcast by the base station into its own utility function and maximize its own utility function to obtain the optimal solution of spectrum resources and price, feed back the optimal solution of spectrum resources and price to the base station, broadcast the optimal solution of price to the user, and the user substitutes the optimal solution of price into his own utility function and maximizes his own utility function to obtain the optimal solution of spectrum resources, and feed back the optimal solution of spectrum resources to the base station and the drone; The iteration module is used to dynamically adjust prices, spectrum resources, commission incentives, and share rewards through overall iteration when the maximum benefits of base stations, drones, and users have not stabilized until base stations, drones, and users obtain maximum benefits.

6. The dynamic spectrum sharing system of the drone network based on the two-stage Stackelberg game according to claim 5 is characterized in that: The iteration module is also used to, after completing a round of iteration, substitute the optimal solution obtained in the previous round into the utility function of the base station in the next round, and conduct a new round of two-stage Stackelberg game, and so on, until the maximum benefits of the base station, drone, and user tend to be stable.

Images