Unmanned aerial vehicle assisted millimeter wave vehicle networking dynamic access optimization method and system based on coalition game

Abstract

The application provides a UAV-assisted millimeter wave V2X dynamic access optimization method and system based on coalition game, and belongs to the technical field of wireless communication, and solves the technical problem of limited network throughput caused by uneven service distribution, communication resource competition and other problems of existing V2X due to millimeter wave being easily blocked and high-speed vehicle movement. The application constructs a vehicle user dynamic access communication system model, discretizes the user access problem in the dynamic environment into an optimization problem in multiple time slots, optimizes the user association and power allocation in each time slot, solves the user association scheme based on the incremental coalition game model under the condition of fixed power allocation vector, preliminarily improves the system throughput, and optimizes the power allocation strategy of each access point based on the current user association scheme by using the Lagrange multiplier method. The application alternately iterates and optimizes the user association and power allocation, improves the network throughput, and significantly reduces the computational complexity.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, specifically to a method and system for dynamic access optimization of unmanned aerial vehicle-to-everything (UAV) assisted millimeter-wave vehicle-to-everything (V2X) networks based on alliance game theory. Background Technology

[0002] Vehicle-to-everything (V2X) communication, as a key component of intelligent transportation systems, plays a vital role in improving traffic efficiency and safety. Millimeter-wave vehicle-to-infrastructure (V2I) communication can provide high data rate transmission to nodes within its coverage area, effectively supporting applications such as traffic management, autonomous driving, and collision prevention. However, due to its high frequency and short wavelength, millimeter waves are easily blocked by vehicles, pedestrians, and buildings in complex urban environments, leading to communication interruptions and poor link stability. Furthermore, the high-speed movement of vehicles causes uneven service distribution and competition for communication resources, further limiting network throughput. These issues have become the technical bottlenecks for the development of millimeter-wave V2X.

[0003] Existing technologies include using drones as relays to enhance the stability of vehicle-to-everything (V2X) links; and proposing low-latency, low-overhead routing protocols to improve the security of drone-assisted vehicle communication. However, current research is mostly based on low-frequency communication, which is insufficient to meet the high-capacity service demands of densely populated user areas.

[0004] Regarding user association schemes, as a key factor affecting the performance of multi-access point communication systems, some studies have systematically explored and verified the effectiveness of millimeter-wave communication user association schemes in improving network performance; other studies have optimized vehicle-to-everything (V2X) association strategies by combining link characteristics. However, these schemes focus on static terrestrial networks, supporting limited system scale and lacking adaptability to dynamic environments, making it difficult to meet the real-time access needs of vehicles.

[0005] To address the energy constraints of drones, existing solutions aim to improve the system's average security rate through power allocation ratios; others employ convex function difference techniques to optimize power allocation and increase speed. However, these solutions are mostly designed for static or low-speed mobile scenarios and struggle to cope with the challenges of frequent user connection switching in vehicle-to-everything (V2X) environments, thus limiting their performance improvement in highly dynamic environments.

[0006] In summary, existing technologies still have many shortcomings: First, user association strategies are mostly targeted at static or low-speed moving scenarios, lacking flexibility and failing to meet the real-time access requirements of drone-assisted millimeter-wave vehicle-to-everything (V2X) networks. Second, fixed power allocation has low spectrum utilization in densely populated vehicle scenarios and ignores the differentiated quality of service requirements of data streams, lacking the ability to adjust in real time according to channel and transmission status, resulting in low resource utilization. Third, in densely populated drone-assisted millimeter-wave V2X networks, the heterogeneity of communication resources between ground and air nodes increases the difficulty of resource allocation, and existing solutions generally have high computational complexity, making it difficult to maintain system performance in fast and dynamic network environments. Summary of the Invention

[0007] Addressing the shortcomings and deficiencies of existing technologies, this invention provides a method and system for dynamic access optimization of UAV-assisted millimeter-wave vehicle-to-everything (V2X) networks. This method and system, based on an incremental alliance game model and employing a joint optimization framework of user association and power allocation, improves resource utilization efficiency and network throughput while significantly reducing computational complexity, thus meeting the access needs of mobile users.

[0008] To achieve the above objectives, the present invention provides the following technical solution: The present invention provides a method and system for dynamic access optimization of unmanned aerial vehicles (UAVs) assisted millimeter-wave vehicle-to-everything (V2X) networks based on alliance game theory, comprising the following steps: S1. Construct a dynamic access communication system for vehicle users, discretize the user access problem in a dynamic environment into an optimization problem in multiple time slots, optimize user association and power allocation in each time slot; divide the time when a vehicle passes through a cell into multiple time slots based on the maximum Doppler frequency shift, and update the location information of UAVs and vehicles in each time slot. S2. Problem modeling: Within each time slot divided in step S1, a mixed integer programming problem is constructed to maximize network capacity, using user association and power allocation as variables. S3. Solve the mixed integer programming problem for network capacity optimization. Decompose the mixed integer programming problem in step S2 into two sub-problems: user association and power allocation. An incremental alliance game model is constructed to solve user association schemes. The model adopts a time slot-level update mechanism and combines it with an incremental user association algorithm. The alliance structure and equilibrium solution of the previous time slot are used as the initial solution, and only the changed users and their associated alliances are partially reconstructed. Based on the current user association scheme, the Lagrange multiplier method is used to solve the power allocation problem of UAVs. By iteratively optimizing user association and power allocation, and combining a time slot-level update mechanism, the resource configuration is continuously optimized until the throughput converges, reaching the optimal solution for network capacity, thereby improving network throughput while significantly reducing computational complexity.

[0009] Preferably, in step S3, the network capacity optimization problem is solved by decomposing the mixed-integer programming problem in step S2 into two sub-problems: user association and power allocation, and then using an alternating iterative optimization method to solve the mixed-integer programming problem. Specifically, this includes: Step S31: Under the condition of fixed power allocation vector, construct an incremental alliance game model, solve the association scheme between users and access points based on the incremental alliance game model, and transform the association problem between users and base stations or drones into a utility maximization alliance game, in which the data streams accessing the base station or the same drone constitute an alliance; adopt a time slot-level update mechanism, combined with an incremental user association algorithm, use the alliance structure and equilibrium solution of the previous time slot as the initial solution, only perform local reconstruction on the changing users and their associated alliances, design an alliance formation function for user association, and optimize the association relationship between users, base stations and drones by iteratively updating the alliance partition until a Nash equilibrium is reached; The incremental user association algorithm is as follows: when the time slot is the first time slot, an initial alliance partition that meets the constraints is generated; When the time slot is not the first time slot, the proportion of users changing in adjacent time slots and the channel changes are detected. If the proportion of users changing in adjacent time slots is ≤50% and the channel correlation is >50%, the alliance structure and equalization solution of the previous time slot are loaded as the initial solution of the current time slot. Only the changing users and their associated alliances are subjected to local reconstruction optimization and switching operations to avoid iterating from zero. If the proportion of users changing in adjacent time slots is >50% or the channel correlation is ≤50%, a new alliance partition that meets the constraints is regenerated and optimized.

[0010] Preferably, the construction of the incremental coalition game model includes the following steps: 1) Participant Set: The data streams requesting network access constitute the participant set of the game, denoted as . ,in Indicates time slot The number of data streams requested for access within the network, which is dynamically updated over time slots, reflects the real-time nature of access requests under the high mobility of connected vehicle users; 2) Alliance and Preference Relationship: The participants were divided into A non-overlapping alliance, that is ,satisfy , ,in Indicates the number of drones. Indicates a set of alliances; For any player Its preference relationship with the alliance is expressed as The utility function of any alliance is expressed as: ,but , indicating player Inclined to join the alliance non-alliance It serves as the basis for switching decisions during partial reconstruction; 3) Profit function: Alliance The benefit is proportional to the total transmission rate of its internal data stream, that is... ,in For the utility calculation factor; for the participant set Arbitrary alliance division The total utility is the sum of the benefits of each alliance, that is... ; 4) Switching operation: Only the changed users and their associated alliances are partially reconstructed, and the preference relationship in step 2) is satisfied; Given a coalition partition When the proportion of users with changes in adjacent time slots is detected to be ≤50% and the channel correlation is >50%, the alliance structure and equalization solution of the previous time slot are loaded as the initial solution for the current time slot. Leave the current alliance And join another alliance At this point, the alliance division was updated to... This enables local reconstruction and optimization, avoiding iteration from scratch.

[0011] Preferably, the method further includes step S32, where, under the condition of a fixed user association vector, the power allocation problem for each UAV based on the current user association scheme is expressed as: , in, Indicates drone The remaining energy; This means for any user; This means for any unmanned aerial vehicle (UAV); Indicates drone Assigned to user The power; This indicates the maximum power of the drone; The power allocation problem of the aforementioned UAV belongs to the convex optimization problem. Based on the current user association scheme, considering power constraints and energy limitations, the Lagrange multiplier is used to construct the Lagrange function, and the power allocation strategy of the UAV is solved by combining the KKT conditions to obtain the optimal power allocation vector.

[0012] Preferably, the method further includes step S33, which involves iteratively optimizing user association and power allocation to solve the mixed integer programming problem, and continuously optimizing the system configuration by combining a time slot-level update mechanism to gradually approach the optimal solution for network capacity until the throughput converges.

[0013] Preferably, in step S1, the vehicle user dynamic access communication system includes a ground base station and Several drone relays work together to provide access services to vehicle users. Both the access point and the vehicle are equipped with directional antennas, establishing millimeter-wave line-of-sight links between users and access points, and between drones and base stations. The millimeter-wave frequency band is divided into... A bandwidth of Sub-channels are used by different access points, and users associated with the same access point use non-orthogonal multiple access. The time taken for a vehicle to pass through the community is divided into segments using a time-segmentation method. Each time slot contains [number] time slots, within which the user location, channel state information, and interference relationships remain unchanged, and the duration of each time slot is:

[0014] in, This represents the maximum Doppler frequency shift.

[0015] Preferably, in step S2, the network capacity is the sum of the data rates of users associated with the base station and the data rates of users associated with the drone, and the specific calculation formula is as follows: For vehicles associated with base stations The received useful power is Based on the decoding principle of non-orthogonal multiple access signals and the characteristics of continuous interference cancellation, its received signal-to-interference-plus-noise ratio (SIR) is:

[0016] in, This indicates the relationship between the user and the base station; Indicates the power of Gaussian white noise; and These represent channel gains greater than The relationship between users and base stations, and their received power; Represents the set of users associated with the base station; For vehicles associated with drone relays The received useful power is Its received signal-to-interference-plus-noise ratio is:

[0017] in, Indicates user and drone The relationship between them; Indicates with drones Associated user set; and These represent channel gains greater than Users and drones The correlation between them and their received power; According to Shannon's capacity formula, the data rates for users associated with base stations and drones are as follows:

[0018] , in, Indicates transceiver efficiency. This indicates the sub-channel bandwidth.

[0019] Therefore, the formula for calculating network capacity is:

[0020] in, and They represent the first Within a time slot, there are base stations and drones. A set of associated links; and They represent the first Data rates of users associated with base stations and drones within each time slot Data rate of associated users.

[0021] Preferably, the optimization variables for user association and power allocation are coupled in the rate expression, forming a mixed-integer programming problem with user association variables and power allocation variables as optimization objectives, and satisfying the following constraints, specifically expressed as follows: , in, Indicates drone The remaining energy; Represents the set of users associated with the base station; Indicates the first A collection of users associated with drones; This indicates the association between a user and an access point; This indicates the relationship between the user and the base station; Indicates user and drone The relationship between them; This means for any user; This means for any unmanned aerial vehicle (UAV); Indicates drone Assigned to user The power; This indicates the maximum power of the drone; constraint This means that each user can only connect to a base station or a specific drone; constraints. and constraints Limit the number of users that a base station and a single drone can serve; constraints and constraints Limit the power allocation at the access point.

[0022] The drone-assisted millimeter-wave vehicle-to-everything (V2X) dynamic access optimization system based on alliance game theory specifically includes: Dynamic time slot division module: used to divide the vehicle service time into multiple time slots and update the location information of the drone and vehicle in each time slot; Problem modeling module: Used to construct mixed-integer programming problems that maximize network capacity with user association and power allocation as variables; Problem-solving module: includes a user association optimization unit, which is used to solve the association scheme between users and access points based on the incremental alliance game model under the condition of fixed power allocation vector, so as to initially improve the system throughput; and a power allocation optimization unit, which is used to solve the power allocation strategy of UAVs based on the current user association scheme under the condition of fixed user association vector, using the Lagrange multiplier method to achieve efficient resource utilization. Alternating Iterative Solution Module: Used to perform alternating iterative optimization of user association and power allocation until the system throughput converges.

[0023] This invention provides a method and system for optimizing dynamic access in UAV-assisted millimeter-wave vehicle-to-everything (V2X) networks based on alliance game theory. It offers the following advantages: (1) The dynamic access optimization method for UAV-assisted millimeter-wave vehicle network based on alliance game theory provided by the present invention designs a joint optimization framework for user association and power allocation, solves the association scheme between users and access points based on the incremental alliance game model, and adjusts the transmission power allocation of each access point based on the Lagrange multiplier method. Finally, it optimizes user association and power allocation through alternating iteration and optimizes resource allocation by combining time slot-level update mechanism, gradually approaching the optimal solution of network capacity. It effectively addresses the problems of uneven service distribution, communication resource competition and millimeter wave susceptibility to blockage caused by high-speed vehicle movement, and significantly improves the network throughput of UAV-assisted millimeter-wave vehicle network.

[0024] Meanwhile, the incremental alliance game model constructed in this invention adopts an incremental user association algorithm. By using the alliance structure and equilibrium solution of the previous time slot as the initial solution, it only performs local reconstruction and optimization on users and their associated alliances that have changed in the network, effectively avoiding redundant calculations that start from scratch in traditional methods. Combined with the power allocation collaborative optimization mechanism, it can accurately adapt to the service capacity constraints of access points, significantly reduce computational complexity while ensuring network performance, and make it adaptable to the fast and dynamic network environment. It can also achieve efficient use of resources and further improve the efficiency of wireless resource utilization.

[0025] (2) The dynamic access optimization method for UAV-assisted millimeter-wave vehicle network based on alliance game theory provided by the present invention divides time slots based on maximum Doppler frequency shift, updates key information such as the location of UAVs and vehicles in each time slot, and iteratively optimizes user association and power allocation in units of time slots, so as to achieve accurate real-time matching of resource allocation and network status. This allows the vehicle user association and UAV power allocation strategies to be flexibly adjusted according to the dynamic changes of channel status, user movement trajectory and service requirements, effectively addressing the problem of frequent changes in network topology caused by high-speed vehicle movement, and meeting the real-time access requirements of UAV-assisted millimeter-wave vehicle network. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the vehicle user dynamic access communication system model in this invention; Figure 2 This is a schematic diagram of the incremental alliance game model framework in this invention. Detailed Implementation

[0027] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0028] The present invention provides a method for dynamic access optimization of unmanned aerial vehicle-to-everything (UAV) assisted millimeter-wave vehicle-to-everything (V2X) networks based on alliance game theory, comprising the following steps: S1. Construct a model for a dynamic access communication system for vehicle users, such as... Figure 1 As shown, the vehicle user dynamic access communication system model includes ground base stations and Several drone relays work together to provide access services to vehicle users. Both the access point and the vehicle are equipped with directional antennas, establishing millimeter-wave line-of-sight links between users and access points, and between drones and base stations. Simultaneously, the millimeter-wave frequency band is divided into... Each bandwidth is Each access point uses a different sub-channel, and users associated with the same access point use non-orthogonal multiple access. The set of users associated with the base station is denoted as . The set of users associated with drones is denoted as ,and , , indicating the first A set of users associated with a drone. For any user binary variables , This indicates its association with the access point. Among them, This indicates its association with the base station.

[0029] The user access problem in a dynamic environment is discretized into an optimization problem within multiple time slots, optimizing user association and power allocation within each time slot. A time-segmentation method is used to divide the vehicle transit time into segments based on the maximum Doppler frequency shift. Each time slot contains [number] time slots, within which the user location, channel state information, and interference relationships remain unchanged. The duration of each time slot is:

[0030] in, This represents the maximum Doppler frequency shift.

[0031] S2. Problem modeling: Within each time slot divided in step S1, a mixed integer programming problem is constructed to maximize network capacity, using user association and power allocation as variables. According to Shannon's capacity formula, the data rates for users associated with base stations and drones are respectively... and ,

[0032]

[0033] in, Indicates transceiver efficiency. This indicates the sub-channel bandwidth.

[0034] Network capacity is the sum of the data rates of users associated with base stations and users associated with drones. The specific calculation formula is as follows: , in, and They represent the first Within a time slot, there are base stations and drones. A set of associated links; and They represent the first Data rates of users associated with base stations and drones within each time slot Data rate of associated users.

[0035] For vehicles associated with base stations The received useful power is Based on the decoding principle of non-orthogonal multiple access signals and the characteristics of continuous interference cancellation, its received signal-to-interference-plus-noise ratio (SIR) is:

[0036] in, This indicates the relationship between the user and the base station; Indicates the power of Gaussian white noise; and These represent channel gains greater than The relationship between users and base stations, and their received power; Represents the set of users associated with the base station; For vehicles associated with drone relays The received useful power is Its received signal-to-interference-plus-noise ratio is:

[0037] in, Indicates user and drone The relationship between them; Indicates with drones Associated user set; and These represent channel gains greater than Users and drones The correlation between them and their received power.

[0038] The optimization variables for user association and power allocation are coupled in the rate expression. This is a mixed-integer programming problem with user association variables and power allocation variables as optimization objectives, and it satisfies the following constraints, specifically expressed as follows:

[0039] in, Indicates drone The remaining energy; Represents the set of users associated with the base station; Indicates the first A collection of users associated with drones; This indicates the association between a user and an access point; This indicates the relationship between the user and the base station; Indicates user and drone The relationship between them; This means for any user; This means for any unmanned aerial vehicle (UAV); This indicates the power allocated to the user by the drone; This indicates the maximum power of the drone; constraint This means that each user can only connect to a base station or a specific drone; constraints. and constraints Limit the number of users that a base station and a single drone can serve; constraints and constraints Limit the power allocation at the access point.

[0040] S3. Problem Solving: Analyze the structural characteristics of the optimization problem, decompose the mixed-integer programming problem in step S2 into two sub-problems: user association and power allocation, and solve the mixed-integer programming problem using an alternating iterative optimization method. Specifically, this includes: S31. Under the condition of a fixed power allocation vector, construct an incremental coalition game model, and solve the association scheme between users and access points based on the incremental coalition game model to initially improve the system throughput; specifically including: like Figure 2 As shown, the problem of associating users with base stations or drones is transformed into a utility-maximizing coalition game, where data flows accessing the base station or the same drone form a coalition. As the number of data flows within the coalition increases, transmission interference increases, and the overall transmission rate decreases. Based on the self-organizing nature of the game, the access link will eventually form... The system consists of several alliances and employs an incremental user association algorithm. By iteratively updating the alliance partitioning, the association between users, base stations, and drones is optimized until a Nash equilibrium is reached. The construction of the incremental coalition game model includes the following steps: 1) Participant Set: The data streams requesting network access constitute the participant set of the game, denoted as . ,in Indicates time slot The number of data streams requested for access within the network, which is dynamically updated over time slots, reflects the real-time nature of access requests under the high mobility of connected vehicle users; 2) Alliance and Preference Relationship: The participants were divided into A non-overlapping alliance, that is ,satisfy , ; For any player Its preference relationship with the alliance is expressed as ,satisfy , indicating player Inclined to join the alliance non-alliance It serves as the basis for switching decisions during partial reconstruction; 3) Profit function: Alliance The benefit is proportional to the total transmission rate of its internal data stream, that is... ,in For the utility calculation factor; for the participant set Arbitrary alliance division The total utility is the sum of the benefits of each alliance, that is... ; 4) Switching operation: Only the changed users and their associated alliances are partially reconstructed, and the preference relationship in step 2) is satisfied; Given a coalition partition When the proportion of users with changes in adjacent time slots is ≤50% and the channel correlation is >50%, the alliance structure and equalization solution of the previous time slot are loaded as the initial solution for the current time slot. Leave the current alliance And join another alliance At this point, the alliance division was updated to... This allows for localized restructuring and optimization, avoiding full iteration.

[0041] Design a coalition formation function for user association, which optimizes the association between users, base stations, and drones by iteratively updating the coalition partition until a Nash equilibrium is reached. The pseudocode for this function is as follows:

[0042] Traditional alliance formation algorithms involve a complete re-game in each time slot, resulting in a computational complexity of O(2^n) (where n is the number of users). To reduce this complexity, the incremental user association algorithm of this invention is as follows: When the time slot is the first time slot, an initial alliance partition that meets the constraints is generated; When the time slot is not the first time slot, the proportion of users changing in adjacent time slots and the channel changes are detected. If the proportion of users changing in adjacent time slots is ≤50% and the channel correlation is >50%, the alliance structure and equalization solution of the previous time slot are loaded as the initial solution of the current time slot. Only the changing users and their associated alliances are subjected to local reconstruction optimization and switching operations to avoid iterating from zero. If the proportion of users changing in adjacent time slots is >50% or the channel correlation is ≤50%, a new alliance partition that meets the constraints is regenerated and optimized.

[0043] The pseudocode for the incremental user association algorithm is as follows:

[0044] S32. Under the condition of fixed user association vector, based on the current user association scheme, considering power constraints and energy limitations, the power allocation strategy of the UAV is solved by the Lagrange multiplier method to achieve efficient utilization of resources. Given a fixed user association vector, the power allocation problem for each UAV can be expressed as:

[0045] in, Indicates drone The remaining energy; This means that for any user; This represents any unmanned aerial vehicle (UAV). Indicates drone Assigned to user The power; Let represent the maximum power of the UAV. The power allocation problem of the UAV mentioned above is a convex optimization problem. Based on the current user association scheme, considering power constraints and energy limitations, a Lagrange function is constructed using Lagrange multipliers, and the power allocation strategy of the UAV is solved by combining KKT conditions to obtain the optimal power allocation vector.

[0046] S33. Alternating iterative optimization of user association and power allocation solves the mixed-integer programming problem, and combined with a time-slot-level update mechanism, continuously optimizes the system configuration, gradually approaching the optimal solution for network capacity until throughput convergence. The pseudocode of the joint optimization algorithm is as follows:

[0047] The drone-assisted millimeter-wave vehicle-to-everything (V2X) dynamic access optimization system based on alliance game theory specifically includes: Dynamic time slot division module: used to divide the vehicle service time into multiple time slots and update the location information of the drone and vehicle in each time slot; Mixed-integer programming modeling module: used to construct mixed-integer programming problems that maximize network capacity with user association and power allocation as variables; Problem-solving module: includes a user association optimization unit, which is used to solve the association scheme between users and access points based on the incremental alliance game model under the condition of fixed power allocation vector, so as to initially improve the system throughput; and a power allocation optimization unit, which is used to solve the power allocation strategy of UAVs based on the current user association scheme under the condition of fixed user association vector, using the Lagrange multiplier method to achieve efficient resource utilization. Alternating Iterative Solution Module: Used to perform alternating iterative optimization of user association and power allocation until the system throughput converges.

[0048] In summary, the UAV-assisted millimeter-wave vehicle-to-everything (V2X) dynamic access optimization method based on coalition game theory of the present invention divides time slots according to the maximum Doppler frequency shift, updates information such as the location of UAVs and vehicles in each time slot, solves the user association scheme based on an incremental coalition game model on a time slot-by-time basis, and optimizes power allocation by combining the Lagrange multiplier method. Through alternating iteration, dynamic collaborative optimization of the two is achieved. It retains the coalition structure and equilibrium solution of the previous time slot and only reconstructs the changed parts locally, which improves network throughput while significantly reducing computational complexity, adapts to access point service constraints, and significantly improves network throughput and wireless resource utilization efficiency while coping with dynamic changes in network topology and channels caused by high-speed vehicle movement and meeting real-time access requirements.

[0049] Based on the above embodiments, the present invention continues to describe in detail the technical features involved therein and the functions and roles of these technical features in the present invention, so as to help those skilled in the art to fully understand the technical solution of the present invention and reproduce it.

[0050] Finally, although this specification describes embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for dynamic access optimization of UAV-assisted millimeter-wave vehicle-to-everything (V2X) networks based on alliance game theory, characterized in that, The steps include the following: S1. Construct a dynamic access communication system for vehicle users, discretize the user access problem in a dynamic environment into an optimization problem in multiple time slots, optimize user association and power allocation in each time slot; divide the time when a vehicle passes through a cell into multiple time slots based on the maximum Doppler frequency shift, and update the location information of UAVs and vehicles in each time slot. S2. Problem modeling: Within each time slot divided in step S1, a mixed integer programming problem is constructed to maximize network capacity, using user association and power allocation as variables. S3. Solve the mixed integer programming problem for network capacity optimization. Decompose the mixed integer programming problem in step S2 into two sub-problems: user association and power allocation. An incremental alliance game model is constructed, and a user association scheme is solved based on the incremental alliance game model. The model adopts a time slot-level update mechanism, combined with an incremental user association algorithm, and uses the alliance structure and equilibrium solution of the previous time slot as the initial solution. Only the changed users and their associated alliances are partially reconstructed. Based on the current user association scheme, the Lagrange multiplier method is used to solve the power allocation problem of UAVs. By iteratively optimizing user association and power allocation, and combining a time slot-level update mechanism, the resource configuration is continuously optimized until the throughput converges, reaching the optimal solution for network capacity, thereby improving network throughput while significantly reducing computational complexity.

2. The method for dynamic access optimization of UAV-assisted millimeter-wave vehicle-to-everything (V2X) networks based on alliance game theory as described in claim 1, characterized in that, In step S3, the network capacity optimization problem is solved by decomposing the mixed-integer programming problem in step S2 into two sub-problems: user association and power allocation. An alternating iterative optimization approach is then used to solve the mixed-integer programming problem, specifically including: Step S31: Under the condition of fixed power allocation vector, construct an incremental alliance game model, solve the association scheme between users and access points based on the incremental alliance game model, and transform the association problem between users and base stations or drones into a utility maximization alliance game, in which the data streams accessing the base station or the same drone constitute an alliance; adopt a time slot-level update mechanism, combined with an incremental user association algorithm, use the alliance structure and equilibrium solution of the previous time slot as the initial solution, only perform local reconstruction on the changing users and their associated alliances, design an alliance formation function for user association, and optimize the association relationship between users, base stations and drones by iteratively updating the alliance partition until a Nash equilibrium is reached; The incremental user association algorithm specifically involves generating an initial alliance partition that satisfies the constraints when the time slot is the first time slot. When the time slot is not the first time slot, the proportion of users changing in adjacent time slots and the channel changes are detected. If the proportion of users changing in adjacent time slots is ≤50% and the channel correlation is >50%, the alliance structure and equalization solution of the previous time slot are loaded as the initial solution of the current time slot. Only the changing users and their associated alliances are subjected to local reconstruction optimization and switching operations to avoid iterating from zero. If the proportion of users changing in adjacent time slots is >50% or the channel correlation is ≤50%, a new alliance partition that meets the constraints is regenerated and optimized.

3. The method for dynamic access optimization of UAV-assisted millimeter-wave vehicle-to-everything (V2X) networks based on alliance game theory as described in claim 2, characterized in that, The construction of the incremental coalition game model specifically includes the following steps: 1) Participant Set: The data streams requesting network access constitute the participant set of the game, denoted as . ,in Indicates time slot The number of data streams requested for access within the network, which is dynamically updated over time slots, reflects the real-time nature of access requests under the high mobility of connected vehicle users; 2) Alliance and Preference Relationship: The participants were divided into A non-overlapping alliance, that is ,satisfy , ,in Indicates the number of drones. Indicates a set of alliances; For any player Its preference relationship with the alliance is expressed as The utility function of any alliance is expressed as: ,but , indicating player Inclined to join the league non-alliance It serves as the basis for switching decisions during partial reconstruction; 3) Profit function: Alliance The benefit is proportional to the total transmission rate of its internal data stream, that is... ,in For the utility calculation factor; for the participant set Arbitrary alliance division The total utility is the sum of the benefits of each alliance, that is... ; 4) Switching operation: Only the changed users and their associated alliances are partially reconstructed, and the preference relationship in step 2) is satisfied; Given a coalition partition When the proportion of users changing in adjacent time slots is ≤50% and the channel correlation is >50%, the alliance structure and equalization solution of the previous time slot are loaded as the initial solution for the current time slot; players Leave the current alliance And join another alliance At this point, the alliance division was updated to... This enables local reconstruction and optimization, avoiding iteration from scratch.

4. The method for dynamic access optimization of UAV-assisted millimeter-wave vehicle-to-everything (V2X) networks based on alliance game theory as described in claim 3, characterized in that, It also includes step S32, under the condition of a fixed user association vector, the power allocation problem for each UAV based on the current user association scheme is expressed as: ， in, Indicates drone The remaining energy; This means for any user; This means for any unmanned aerial vehicle (UAV); Indicates drone Assigned to user The power; This indicates the maximum power of the drone; The power allocation problem of the aforementioned UAV belongs to the convex optimization problem. Based on the current user association scheme, considering power constraints and energy limitations, the Lagrange multiplier is used to construct the Lagrange function, and the power allocation strategy of the UAV is solved by combining the KKT conditions to obtain the optimal power allocation vector.

5. The method for dynamic access optimization of UAV-assisted millimeter-wave vehicle-to-everything (V2X) networks based on alliance game theory as described in claim 4, characterized in that, It also includes step S33, which iteratively optimizes user association and power allocation to solve the mixed integer programming problem, and continuously optimizes the system configuration by combining a time slot-level update mechanism, gradually approaching the optimal solution for network capacity until the throughput converges.

6. The method for dynamic access optimization of UAV-assisted millimeter-wave vehicle-to-everything (V2X) networks based on alliance game theory as described in claim 1, characterized in that, In step S1, the vehicle user dynamic access communication system includes a ground base station and Several drone relays work together to provide access services to vehicle users. Both the access point and the vehicle are equipped with directional antennas, establishing millimeter-wave line-of-sight links between users and access points, and between drones and base stations. The millimeter-wave frequency band is divided into... A bandwidth of Sub-channels are used by different access points, and users associated with the same access point use non-orthogonal multiple access. The time taken for a vehicle to pass through the community is divided into segments using a time-segmentation method. Each time slot contains [number] time slots, within which the user location, channel state information, and interference relationships remain unchanged, and the duration of each time slot is: ， in, This represents the maximum Doppler frequency shift.

7. The method for dynamic access optimization of UAV-assisted millimeter-wave vehicle-to-everything (V2X) networks based on alliance game theory as described in claim 1, characterized in that, In step S2, the network capacity is the sum of the data rates of users associated with the base station and users associated with the drone. The specific calculation formula is as follows: For vehicles associated with base stations The received useful power is Based on the decoding principle of non-orthogonal multiple access signals and the characteristics of continuous interference cancellation, its received signal-to-interference-plus-noise ratio (SIR) is: ， in, This indicates the relationship between the user and the base station; Indicates the power of Gaussian white noise; and These represent channel gains greater than The relationship between users and base stations, and their received power; Represents the set of users associated with the base station; For vehicles associated with drone relays The received useful power is Its received signal-to-interference-plus-noise ratio is: ， in, Indicates user and drone The relationship between them; Indicates with drones Associated user set; and These represent users with channel gain greater than [value missing]. With drones The correlation between the base station and the received power; according to Shannon's capacity formula, the data rates of users associated with the base station and the drone are respectively: ; ， in, Indicates transceiver efficiency. Indicates the sub-channel bandwidth; Therefore, the formula for calculating network capacity is: ， ， in, and They represent the first Within a time slot, there are base stations and drones. A set of associated links; and They represent the first Data rates of users associated with base stations and drones within each time slot Data rate of associated users.

8. The method for dynamic access optimization of UAV-assisted millimeter-wave vehicle-to-everything (V2X) networks based on alliance game theory as described in claim 7, characterized in that, The optimization variables for user association and power allocation are coupled in the rate expression. This is a mixed-integer programming problem with user association variables and power allocation variables as optimization objectives, and it satisfies the following constraints, specifically expressed as follows: ， in, Indicates drone The remaining energy; Represents the set of users associated with the base station; Indicates the first A collection of users associated with drones; This indicates the association between a user and an access point; This indicates the relationship between the user and the base station; Indicates user and drone The relationship between them; This means for any user; This means for any unmanned aerial vehicle (UAV); Indicates drone Assigned to user The power; This indicates the maximum power of the drone; constraint This means that each user can only connect to a base station or a specific drone; constraints. and constraints Limit the number of users that a base station and a single drone can serve; constraints and constraints Limit power allocation at access points.

9. A drone-assisted millimeter-wave vehicle-to-everything (V2X) dynamic access optimization system based on alliance game theory, characterized in that: The method for dynamic access optimization of UAV-assisted millimeter-wave vehicle-to-everything (V2X) based on alliance game theory as described in any one of claims 1-8 specifically includes: Dynamic time slot division module: used to divide the vehicle service time into multiple time slots and update the location information of the drone and vehicle in each time slot; Problem modeling module: Used to construct mixed-integer programming problems that maximize network capacity with user association and power allocation as variables; Problem-solving module: includes a user association optimization unit, which is used to solve the association scheme between users and access points based on the incremental alliance game model under the condition of fixed power allocation vector, so as to initially improve the system throughput; and a power allocation optimization unit, which is used to solve the power allocation strategy of UAVs based on the current user association scheme under the condition of fixed user association vector, using the Lagrange multiplier method to achieve efficient resource utilization. Alternating Iterative Solution Module: Used to perform alternating iterative optimization of user association and power allocation until the system throughput converges.

Images