Bandwidth allocation and user grouping method and system based on game theory

Abstract

The present invention discloses a bandwidth allocation and user grouping method and system based on game theory. A first objective function is established with the goal of minimizing the total amount of remaining data of the drone; a drone bandwidth allocation model is obtained, which is established with the second objective function of maximizing the amount of data collected by the drone, and the drone bandwidth allocation model is solved to obtain an optimal bandwidth allocation scheme; based on the optimal bandwidth allocation scheme, the first objective function is converted into a user grouping model within each time slot; the user grouping model is solved using an alliance formation game algorithm to obtain stable non-overlapping user groupings. The user groupings are further optimized using an overlapping alliance formation game algorithm. The present invention can obtain a stable and optimal grouping scheme through alliance formation game, and then use overlapping alliance formation game to allow users to join multiple groups to further improve performance. The present invention reduces the overall computational complexity and avoids the accumulation of data at the drone.

Description

Technical Field

[0001] The invention relates to a bandwidth allocation and user grouping method and system based on game theory, and belongs to the field of information communication. Background Art

[0002] Drones, as aerial base stations, can achieve wide-area coverage and integrated ground-to-ground communications, making them a key development direction for 5G and the future 6G. Mobile edge computing allows computing servers to be deployed anywhere at the edge of the network, shifting computing power from the core network to the edge of the access network. This significantly improves the quality of communication, computing, caching, and control. Drones, with their high mobility and flexibility, can provide seamless, reliable, and low-latency communication services. The integration of drones and mobile edge computing is an inevitable trend. Drone-based edge computing networks can provide reliable, cost-effective, low-latency, and efficient wireless communication services.

[0003] Existing technologies primarily focus on drone energy consumption and spectrum efficiency, but the mismatch between drone computing power and the volume of data collected by data nodes within their coverage area remains unresolved. This means the volume of data collected within a drone's coverage area may exceed its computing capacity. Furthermore, traditional centralized solutions struggle to access the massive number of data nodes. Summary of the Invention

[0004] The purpose of the present invention is to propose a bandwidth allocation and user grouping method and system based on game theory with low computational complexity, so as to match the amount of user data with the computing power of drones, avoid data accumulation in drones, and minimize the amount of data at drones in drone edge computing networks.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] In one aspect, a bandwidth allocation and user grouping method based on game theory includes:

[0007] The first objective function is established with the goal of minimizing the total amount of remaining data of the drone and the drone bandwidth and drone processing capacity as constraints;

[0008] Obtaining a UAV bandwidth allocation model established with maximizing UAV-collected data as a second objective function, solving the UAV bandwidth allocation model to obtain an optimal bandwidth allocation solution;

[0009] Based on the optimal bandwidth allocation solution, converting the first objective function into a user grouping model within each time slot;

[0010] The user grouping model is solved using a coalition formation game algorithm to obtain stable non-overlapping user groups.

[0011] Furthermore, the first objective function is:

[0012]

[0013] Where A is N u ×N v The matrix, N u and N v They are the number of data nodes distributed in the UAV Internet of Things and the number of UAVs with edge computing capabilities. is a collection of data nodes, For the collection of drones, a ij (t) is the element in row i and column j of A, a ij (t) is a binary indicator variable, a ij (t) = 1 indicates a data node Upload to drone a ij (t) = 0 means Not uploading to drone B=[b1,b2,b3,…,b Nv ], b j =[b 1j (t),b 2j (t),b 3j (t),…,b Nuj (t)],b ij (t) is the data node Transmit to drone The bandwidth of the time slot is T; T is the maximum number of time slots, D j (T) is a drone The amount of data remaining in the Tth time slot; B t The bandwidth for each drone; d ij (t) is the data node With drones The distance between th is the distance threshold between the data node and the drone; D j (t-1) is the drone The amount of data remaining in time slot t-1; For data nodes Transmit to the UAV at time slot t The maximum amount of data, in For data nodes The amount of data remaining in time slot t, R ij (t) is the data node Transmit to drone The achievable rate, where hij (t) and Data nodes Transmit to drone The channel gain and transmission power when is the noise power, in For data nodes The maximum transmission power, τ is the time slot length; C j For drones The computing power of Ca j The cache stack size of the drone.

[0014] Furthermore, the UAV bandwidth allocation model is:

[0015]

[0016] Furthermore, the Lagrange multiplier method is used to solve the UAV bandwidth allocation model, specifically including:

[0017] The UAV bandwidth allocation model is transformed into a non-negative dual variable η j and the Lagrange multiplier λ j =[λ 1j ,λ 2j ,λ 3j ,…,λ Nuj ]’s Lagrangian function:

[0018]

[0019] According to the KKT condition, The optimal solution of is given by the following conditions:

[0020]

[0021] From formula (9), we can get η j =log2(1+ρ ij (t))-ρ ij (t) / (ln2(1+ρ ij (t))), Therefore, the optimal bandwidth allocation solution is:

[0022]

[0023] Furthermore, the user grouping model in each time slot is:

[0024]

[0025] Among them, C j (t) is the UAV in time slot t Real-time computing capabilities, The drone in time slot t The matching degree between the amount of collected data and real-time computing capabilities.

[0026] Furthermore, the method of solving the user grouping model by utilizing the alliance formation game algorithm includes:

[0027] Step 61: The data node randomly selects a drone to access and obtains the initial group in Indicates to drone The set of data nodes that upload data, set the corresponding indicator vector a ij (t) = 1;

[0028] Step 62: Randomly select a data node and its grouping Calculate its utility function V c ;

[0029] Step 63, data node Join adjacent groups respectively in For data nodes to drones The distance between the two points is calculated, and the utility function is calculated. The set of utility functions is set as vector V p , V p Including data nodes Add all utility function values ​​of its adjacent groups respectively;

[0030] Step 64, find vector V p The maximum value of V p =maxV p , and the maximum value V p The number of the corresponding set

[0031] Step 65, if V p >V c , then the data node Leave the original group Join Group And set the corresponding indicator variables

[0032] Repeat steps 62 to 65 until the grouping does not change any more, and then output the grouping.

[0033] Furthermore, the bandwidth allocation and user grouping method based on game theory further includes: using overlapping coalitions to form a game algorithm, and continuously optimizing the stable non-overlapping user groupings to obtain optimal user groupings.

[0034] Furthermore, the game algorithm formed by utilizing overlapping alliances to continuously optimize the stable non-overlapping user groups includes:

[0035] For the stable non-overlapping user groups Randomly select a data node And the N it adds i Groups And find the adjacent group that does not include it Among them, data nodes With drones The distance is less than or equal to the distance threshold d th ,Right now

[0036] If N i <L max , L max The maximum number of groups that a data node can join, and the number of all groups including data nodes is calculated. The sum of the amount of data collected by the UAV in time slot t and the matching degree of real-time computing capability is N i Groups All data nodes use power and The difference V c , and adjacent drones The matching degree between the amount of collected data in time slot t and the real-time computing capability is in the set All data nodes in the and The difference V pi ,in

[0037] V for all vector collections pi The set is V p =[V p1 ,V p2 ,V p3 ,…], and find the vector V p The minimum value of V p =minV p , and the minimum value V p The number of the corresponding set

[0038] If V p >V c , then the data node Join Group And set the corresponding indicator variables gather The power of the data nodes in

[0039] Repeat the above steps until the grouping no longer changes, and then output the grouping.

[0040] On the other hand, a bandwidth allocation and user grouping system based on game theory includes:

[0041] a first objective function establishing module configured to establish the first objective function with minimization of the total amount of remaining data of the UAV as a goal and with the bandwidth of the UAV and the processing capacity of the UAV as constraints;

[0042] a UAV bandwidth optimization module configured to obtain a UAV bandwidth allocation model established with maximizing UAV-collected data as a second objective function, solve the UAV bandwidth allocation model, and obtain an optimal bandwidth allocation solution;

[0043] A user grouping model establishing module configured to convert the first objective function into a user grouping model within each time slot based on the optimal bandwidth allocation solution;

[0044] The user grouping solution module uses a coalition formation game algorithm to solve the user grouping model and obtain stable non-overlapping user groups.

[0045] Furthermore, the bandwidth allocation and user grouping system based on game theory further includes:

[0046] The user grouping optimization module is configured to use the overlapping alliance to form a game algorithm to continuously optimize the stable non-overlapping user groupings to obtain the optimal user groupings.

[0047] The present invention also provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions that, when executed by a computing device, enable the computing device to perform the aforementioned game theory-based bandwidth allocation and user grouping method.

[0048] The present invention also provides a computing device, comprising:

[0049] One or more processors, one or more memories, and one or more programs, wherein the one or more programs are stored in the one or more memories and are configured to be executed by the one or more processors, and the one or more programs include methods for executing the aforementioned game theory-based bandwidth allocation and user grouping method.

[0050] The beneficial technical effects achieved by the present invention are:

[0051] (1) The present invention uses the Lagrange multiplier method to obtain a closed-form expression for the dynamic allocation bandwidth of each UAV. This expression is only related to the channel gain of the UAV, has low computational complexity, and is easy to implement.

[0052] (2) The present invention uses alliance formation game, where users exchange information with each other and distribute the exchange groups, ultimately reaching a stable and optimal grouping solution. Furthermore, the overlapping alliance formation game is used to allow users to join multiple groups, further improving performance.

[0053] (3) The present invention converts a long-time-slot problem into a single-time-slot problem as a whole, reducing the overall computational complexity. The method is simpler and can process the collected data more quickly than other criteria and schemes, thus avoiding the accumulation of data at the drone. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flow chart of the method of the present invention;

[0055] Figure 2 It is a flow chart of the alliance formation game algorithm;

[0056] Figure 3 It is a flowchart of the game algorithm for overlapping alliance formation;

[0057] Figure 4 It is a convergence diagram of the coalition formation game algorithm;

[0058] Figure 5 It is a diagram of the convergence of the overlapping coalition formation game algorithm;

[0059] Figure 6 Schematic diagram of the performance of the method of the present invention. DETAILED DESCRIPTION

[0060] The present invention will be further described below in conjunction with specific examples. The following examples are only used to more clearly illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention.

[0061] In the drone-based Internet of Things, there are N u data nodes and N v A drone with edge computing capabilities. The set of data nodes is The collection of drones is The computing power of the drone is Among them, C j For drones Computing power, measured in bits. Traditional centralized grouping methods are insufficient to support the simultaneous access of large numbers of users, and fixed bandwidth allocation lacks flexibility. Game theory, due to its distributed autonomy and stable results, has been widely used in the communications field.

[0062] To this end, the present invention proposes a bandwidth allocation and user grouping method based on game theory for UAV Internet of Things. Figure 1 As shown, the method specifically includes:

[0063] Step S1, establishing a first objective function with minimizing the total amount of remaining data of the drone as the goal and the drone bandwidth and drone processing capacity as constraints;

[0064] Transmission model: Data nodes use FDMA to transmit data to drones, and the bandwidth of each drone is B t , the bandwidth that a data node can occupy is arbitrary. Then the data node Transmit to drone The achievable rate is:

[0065]

[0066] Among them, h ij (t), b ij (t) and Data nodes Transmit to drone Channel gain, bandwidth and transmission power when is the noise power.

[0067] Therefore, the data node Transmit to drone The maximum amount of data is:

[0068]

[0069] in, For data nodes The amount of data remaining in time slot t, in For drones The amount of data remaining in time slot t-1, It is a data node The data generated or collected in time slot t, τ is the time slot length, when t = 1, a ij (t) is a binary indicator variable, a ij (t) = 1 indicates a data node Upload to drone On the contrary, a ij (t) = 0 means Not uploading to drone

[0070] In order to ensure the fairness of data transmission among data nodes, the transmission power of drones is evenly distributed, i.e. in, For data nodes The maximum transmission power.

[0071] Considering the limited communication distance of data nodes, in order to ensure reliable communication, data nodes With drones The distance d ij (t) should be less than the distance threshold d th In addition, the amount of data at the drone should be smaller than its processing capacity and the size of its cache stack, that is:

[0072] d ij (t)≤d th , (3)

[0073]

[0074] Among them, Ca j is the cache stack size of the drone, D j (t) is a drone The amount of data remaining in time slot t is uniformly distributed. j The expression of (t) is:

[0075]

[0076] The present invention minimizes the amount of residual data at a UAV for a long time by optimizing bandwidth and user grouping.

[0077] The optimization objective function is established with the goal of minimizing the total amount of remaining data of the drone:

[0078]

[0079] Where A is N u ×N v The matrix, a ij (t) is the element in row i and column j of A, T is the maximum number of time slots, D j (T) is a drone The amount of data remaining in the Tth time slot; B t is the bandwidth of each drone. C1 and C2 are bandwidth constraints, C3 is the distance constraint, and C4 is the drone processing capacity constraint.

[0080] Problem P is based on the constraints of bandwidth and UAV processing capacity, and aims to minimize the amount of remaining data at the UAV by optimizing the bandwidth and grouping of users.

[0081] Step S2: obtaining a UAV bandwidth allocation model established with maximizing the amount of data collected by the UAV as a second objective function, solving the UAV bandwidth allocation model, and obtaining an optimal power allocation solution;

[0082] To solve problem P, we first need to solve the bandwidth allocation problem of drones. For drones, the purpose is to collect as much data as possible, so we establish a drone bandwidth allocation model:

[0083]

[0084] It can be seen that problem P1 is a convex function and can be solved using the Lagrange multiplier method. Therefore, it has a non-negative dual variable η j and Lagrange multipliers The Lagrangian function is:

[0085]

[0086] According to the KKT condition, The optimal solution of is given by the following conditions:

[0087]

[0088] From formula (9), we can get η j =log2(1+ρ ij (t))-ρ ij (t) / (ln2(1+ρ ij (t))),

[0089] Therefore, the optimal bandwidth allocation solution is:

[0090]

[0091] Step S3: based on the optimal bandwidth allocation solution, converting the first objective function into a user grouping model within each time slot;

[0092] Because D j (t) Only with D j (t-1) and The original problem P can be transformed into the user grouping problem in each time slot, that is:

[0093]

[0094] Among them, C j (t) is the UAV in time slot t Real-time computing capabilities, The drone is defined as the time slot t The matching degree between the amount of collected data and real-time computing capabilities.

[0095] Step S4: using the alliance formation game algorithm to solve the user grouping model and obtain stable non-overlapping user groups.

[0096] Since the alliance formation game is synergistic, a sub-network can be formed in a low-complexity distributed manner. Using the alliance formation game algorithm, a non-overlapping user group can be obtained. Right now: Indicates to drone A collection of data nodes that upload data.

[0097] Defining data nodes The utility function is:

[0098]

[0099] in, For data nodes In group The utility function of When the time slot is t The data node in the drone The amount of data transferred, When the time slot is t The data node in the drone The amount of data transferred, C j1 (t) is the time slot of the drone Real-time computing capabilities.

[0100] Defining data nodes The grouping change criteria are:

[0101]

[0102] in, When the time slot is t The data node in the drone The amount of data transferred, When the time slot is t The data node in the drone The amount of data transferred, t time slot drone This formula indicates that the data node is Leave the original group Join an existing group

[0103] like Figure 2As shown, the alliance formation game (Algorithm 1) is used to form a stable non-overlapping user group. The algorithm steps are as follows:

[0104] 1. Initialization: The data node randomly selects a drone to access and obtains the initial group Set the corresponding indicator vector to 1, that is, a ij (t) = 1;

[0105] 2. Cycle;

[0106] 3. Randomly select a data node and its grouping (Data Node ), calculate its utility function;

[0107] 4. Data Node Join adjacent groups respectively (Data Node ), in, For data nodes to drones The distance between the two points is calculated, and the utility function is calculated. The set of utility functions is set as vector V p , V p Including data nodes Add all utility function values ​​of its adjacent groups respectively;

[0108] 5. Find vector V p The maximum value of V p =maxV p , and the maximum value V p The number of the corresponding set

[0109] 6. If V p >V c ;

[0110] 7. Data nodes Leave the original group Join Group And set the corresponding indicator variables

[0111] 8. End if;

[0112] 9. If the grouping does not change, output the grouping. Otherwise, return to step 2.

[0113] In a further embodiment, the method of the present invention further comprises:

[0114] Step S5: Using the overlapping alliance formation game algorithm, the stable non-overlapping user groupings are further optimized to obtain the optimal user groupings.

[0115] In order to further adapt the computing power of drones to the amount of data collected, this paper proposes a grouping method based on the overlapping alliance formation game (Algorithm 2), such as Figure 3 As shown, the specific algorithm flow is as follows:

[0116] 1. Initialization: Get a non-overlapping group of data nodes through Algorithm 1

[0117] 2. Cycle;

[0118] 3. Randomly select a data node And the N it adds i Groups (Data Node ), and find the adjacent group that does not include it (Data Node ),

[0119] 4. If N i <L max , L max The maximum number of groups that a data node can join;

[0120] 5. Calculate all data nodes at this time The sum of the amount of data collected by the UAV in time slot t and the matching degree of real-time computing capability is N i Groups All data nodes use power and The difference:

[0121]

[0122] and adjacent drones The matching degree between the amount of collected data in time slot t and the real-time computing capability is in the set All data nodes in the and The difference:

[0123] in,

[0124] 6. V of the set of all vectors pi The set is V p =[V p1 ,V p2 ,V p3 ,…], and find the vector Vp The minimum value of V p =minV p , and the minimum value V p The number of the corresponding set

[0125] 7. If V p >V c ;

[0126] 8. Data nodes Join Group And set the corresponding indicator variables gather The power of the data nodes in is set as:

[0127] 9. End if;

[0128] 10. End if;

[0129] 11. If the grouping does not change, output the grouping. Otherwise, return to step 2.

[0130] In order to demonstrate the performance advantages of the method of the present invention, the following simulation verification was performed. The simulation parameters were set as follows: h ij (t) obeys the Rayleigh distribution with mean 1, Obey the uniform distribution of U(200,400) bits, C j Obey the uniform distribution of U(350,450) bits, B t =1MHz, N u =20,N v =10.

[0131] Figure 4 and Figure 5 The convergence of the matching degree between the amount of collected data in a time slot and the real-time computing power was simulated. As shown in the figure, both Algorithms 1 and 2 have good convergence. As the number of iterations increases, the matching degree of Algorithm 1 gradually increases and eventually stabilizes. When the threshold distance is larger, the user has more groups to choose from, and can select better groups to improve the matching degree. Therefore, when the distance threshold changes from 5km to 10km, the matching degree of Algorithm 1 is higher. Compared with fixed bandwidth allocation schemes, the proposed bandwidth allocation scheme is more flexible. Therefore, the bandwidth allocation scheme of the present invention has a better matching degree than traditional bandwidth allocation schemes.

[0132] Figure 5 It shows the convergence performance of Algorithm 2. When the number of iterations increases, L max = 1 is stable, and Lmax =2 and L max =4 The matching degree between real-time computing power and the amount of collected data increases within a limited number of iterations and then remains stable. This is because when the number of groups that users can join increases, all data nodes have more drones uploading data. The group selection strategy of data nodes is more flexible and can achieve a higher matching degree. Therefore, L max =4 is better than L max =2 and L max =1.

[0133] Figure 6 The following is a performance diagram of the method of the present invention, comparing it with the Pareto criterion, the selfish criterion, and the recent criterion. As the total number of time slots increases, the data volume of the selfish, Pareto, and recent criteria increases dramatically, while the data volume of the method of the present invention remains relatively stable. Furthermore, the more groups a data node joins, the less data volume there is. This is because the grouping change criterion of the present invention better matches the user's real-time computing power with the amount of collected data, reducing data accumulation at the drone. Other solutions fail to process the collected data in a timely manner, resulting in data accumulation.

[0134] In another embodiment, a bandwidth allocation and user grouping system based on game theory includes:

[0135] a first objective function establishing module configured to establish the first objective function with minimization of the total amount of remaining data of the UAV as a goal and with the bandwidth of the UAV and the processing capacity of the UAV as constraints;

[0136] a UAV bandwidth optimization module configured to obtain a UAV bandwidth allocation model established with maximizing UAV-collected data as a second objective function, solve the UAV bandwidth allocation model, and obtain an optimal bandwidth allocation solution;

[0137] A user grouping model establishing module configured to convert the first objective function into a user grouping model within each time slot based on the optimal bandwidth allocation solution;

[0138] The user grouping solution module uses a coalition formation game algorithm to solve the user grouping model and obtain stable non-overlapping user groups.

[0139] Furthermore, the bandwidth allocation and user grouping system based on game theory further includes:

[0140] The user grouping optimization module is configured to use the overlapping alliance to form a game algorithm to continuously optimize the stable non-overlapping user groupings to obtain the optimal user groupings.

[0141] The present invention also provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions that, when executed by a computing device, enable the computing device to perform the aforementioned game theory-based bandwidth allocation and user grouping method.

[0142] The present invention also provides a computing device, comprising:

[0143] One or more processors, one or more memories, and one or more programs, wherein the one or more programs are stored in the one or more memories and are configured to be executed by the one or more processors, and the one or more programs include methods for executing the aforementioned game theory-based bandwidth allocation and user grouping method.

[0144] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0145] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0146] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0147] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0148] The present invention has been disclosed above with preferred embodiments, which are not intended to limit the present invention. Any technical solutions obtained by adopting equivalent replacement or equivalent transformation solutions fall within the protection scope of the present invention.

Claims

1. A bandwidth allocation and user grouping method based on game theory, characterized in that: include: The first objective function is established with the goal of minimizing the total amount of remaining data of the drone and the drone bandwidth and drone processing capacity as constraints; Obtaining a UAV bandwidth allocation model established with maximizing UAV-collected data as a second objective function, solving the UAV bandwidth allocation model to obtain an optimal bandwidth allocation solution; Based on the optimal bandwidth allocation solution, converting the first objective function into a user grouping model within each time slot; Using the alliance formation game algorithm, the user grouping model is solved to obtain stable non-overlapping user groups; The optimal bandwidth allocation scheme is: ； in, yes The element in row i and column j in yes The matrix, and They are the number of data nodes distributed in the UAV Internet of Things and the number of UAVs with edge computing capabilities. is a collection of data nodes, , For the collection of drones, , is a binary indicator variable, Represents a data node Upload to drone , express Not uploading to drone ; For data nodes Transmit to drone bandwidth when and Data nodes Transmit to drone Channel gain and transmission power when Bandwidth for each drone.

2. The bandwidth allocation and user grouping method based on game theory according to claim 1, characterized in that: The first objective function is: (6) in, yes The matrix, and They are the number of data nodes distributed in the UAV Internet of Things and the number of UAVs with edge computing capabilities. is a collection of data nodes, , For the collection of drones, ; yes The element in row i and column j in is a binary indicator variable, Represents a data node Upload to drone , express Not uploading to drone ; , , For data nodes Transmit to drone The bandwidth of the time slot is 2000; T is the maximum number of time slots, For drones The amount of data remaining in the Tth time slot; bandwidth for each drone; For data nodes With drones The distance between is the distance threshold between the data node and the drone; For drones exist -1 The amount of data remaining in the time slot; For data nodes exist Time slot transmission to drone The maximum amount of data, in For data nodes The amount of data remaining in time slot t, It is a data node Transmit to drone The achievable rate, ,in and Data nodes Transmit to drone The channel gain and transmission power when is the noise power, ,in For data nodes The maximum transmission power, is the time slot length; For drones computing power, The cache stack size of the drone.

3. The bandwidth allocation and user grouping method based on game theory according to claim 2, characterized in that: The UAV bandwidth allocation model is: (7) 4. The bandwidth allocation and user grouping method based on game theory according to claim 3, characterized in that: The Lagrange multiplier method is used to solve the UAV bandwidth allocation model, specifically including: Transform the UAV bandwidth allocation model into a model with non-negative dual variables and Lagrange multipliers The Lagrangian function of : (8) According to the KKT condition, The optimal solution of is given by the following conditions: (9) From formula (9), we can get: , , therefore, the optimal bandwidth allocation scheme is: 。 (10) 5. The bandwidth allocation and user grouping method based on game theory according to claim 4, characterized in that: The user grouping model in each time slot is: (11) in, The drone in time slot t Real-time computing capabilities, The drone in time slot t The matching degree between the amount of collected data and real-time computing capabilities.

6. The bandwidth allocation and user grouping method based on game theory according to claim 5, characterized in that: The method of solving the user grouping model by utilizing the alliance formation game algorithm includes: Step 61: The data node randomly selects a drone to access and obtains the initial group ,in Indicates to drone The set of data nodes that upload data and set the corresponding indicator vector ; Step 62: Randomly select a data node , and its grouping , calculate its utility function ; Step 63, data node Join adjacent groups respectively , ,in For data nodes to drones The distance and calculate the utility function at this time, setting the set of utility functions as vector , Including data nodes Add all utility function values ​​of its adjacent groups respectively; Step 64, find the vector The maximum value , and the maximum value The corresponding set number ; Step 65, if , then the data node Leave the original group Join Group , and set the corresponding indicator variables ; Repeat steps 62 to 65 until the grouping does not change any more, and then output the grouping.

7. The bandwidth allocation and user grouping method based on game theory according to claim 6, characterized in that: Also includes: The overlapping alliance is used to form a game algorithm, and the stable non-overlapping user grouping is further optimized to obtain the optimal user grouping.

8. The bandwidth allocation and user grouping method based on game theory according to claim 7, characterized in that: The method of forming a game algorithm by utilizing overlapping alliances and continuously optimizing the stable non-overlapping user groups includes: For the stable non-overlapping user groups , randomly select a data node And it added Groups , and find the adjacent group that does not include it , where data nodes With drones The distance is less than or equal to the distance threshold ; like , The maximum number of groups that a data node can join, and the number of all groups including data nodes is calculated. The sum of the amount of data collected by the drone in time slot t and the matching degree of real-time computing capability is Groups All data nodes use power and The difference , and adjacent drones The matching degree between the amount of collected data in time slot t and the real-time computing capability is in the set All data nodes in the and The difference ,in ; All vector collections The collection is , and find the vector Minimum value of , and the minimum The corresponding set number ; like , then the data node Join Group , and set the corresponding indicator variables ,gather The power of the data nodes in ; Repeat the above steps until the grouping no longer changes, and then output the grouping.

9. A bandwidth allocation and user grouping system based on game theory, characterized in that: include: a first objective function establishing module configured to establish the first objective function with minimization of the total amount of remaining data of the UAV as a goal and with the bandwidth of the UAV and the processing capacity of the UAV as constraints; a UAV bandwidth optimization module configured to obtain a UAV bandwidth allocation model established with maximizing UAV-collected data as a second objective function, solve the UAV bandwidth allocation model, and obtain an optimal bandwidth allocation solution; A user grouping model establishing module configured to convert the first objective function into a user grouping model within each time slot based on the optimal bandwidth allocation solution; A user grouping solution module, which uses a coalition formation game algorithm to solve the user grouping model and obtain stable non-overlapping user groups; The optimal bandwidth allocation scheme is: ； in, yes The element in row i and column j in yes The matrix, and They are the number of data nodes distributed in the UAV Internet of Things and the number of UAVs with edge computing capabilities. is a collection of data nodes, , For the collection of drones, , is a binary indicator variable, Represents a data node Upload to drone , express Not uploading to drone ; For data nodes Transmit to drone bandwidth when and Data nodes Transmit to drone Channel gain and transmission power when Bandwidth for each drone.

10. A bandwidth allocation and user grouping system based on game theory according to claim 9, characterized in that: Also includes: The user grouping optimization module is configured to use the overlapping alliance to form a game algorithm to continuously optimize the stable non-overlapping user groupings to obtain the optimal user groupings.

Images