A Method for UAV Swarm Task Scheduling and Channel Allocation Based on Potential Game

Abstract

The present invention discloses a method for unmanned aerial vehicle (UAV) swarm task scheduling and channel allocation based on potential game, comprising the following steps: (1) In a UAV swarm-assisted ground communication network, establish a scenario for data transmission from the UAV swarm to ground users; (2) Model the UAV swarm task scheduling and channel allocation problem as a trajectory optimization and channel access game model based on potential game, wherein the game participants are the set of UAVs; (3) Use a user scheduling and channel joint selection method based on best response to solve the trajectory optimization and channel access game model, so as to maximize the total utility of all ground users served by each UAV, and complete the UAV swarm task scheduling and channel allocation; The present invention effectively improves the task execution utility of UAVs.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and particularly to a method for unmanned aerial vehicle (UAV) swarm task scheduling and channel allocation based on potential game. Background Art

[0002] The rapid development of terrestrial wireless communication networks can provide good communication services for terrestrial users and meet their daily communication needs. However, with the explosive growth of terrestrial wireless devices, the terrestrial communication network shows a trend of densification. At the same time, terrestrial wireless communication is usually affected by buildings, resulting in severe multipath fading of communication links and making it difficult to provide reliable communication services for users. UAVs, with their advantages such as high dynamicity, dynamic configuration and deployment, have been widely used in UAV-assisted terrestrial communication networks. UAVs can be used as aerial base stations to increase the possibility of line-of-sight transmission with terrestrial users and enable them to obtain better transmission performance.

[0003] Currently, in the research on air-ground integrated networks, existing work mainly focuses on improving aspects such as the overall network rate by jointly optimizing the position deployment of UAV swarms, channel access, power control, etc. However, in most studies, UAV task scheduling only considers user services at specific positions, that is, once the UAV positions are optimized, they are fixed and only serve specific users. In addition, in the studies considering UAV trajectory optimization, most studies assume that there are sufficient channels in the network and there is no mutual interference between users. However, with the increase in terrestrial users and the dynamic movement of UAVs, the dynamic interference of user links has become very serious. Summary of the Invention

[0004] Object of the Invention: The object of the present invention is to provide a method for UAV swarm task scheduling and channel allocation based on potential game to solve the improvement of the ground communication transmission performance of UAVs in air-ground integrated networks.

[0005] Technical Solution: A method for UAV swarm task scheduling and channel allocation based on potential game according to the present invention includes the following steps:

[0006] (1) In a UAV swarm-assisted terrestrial communication network, establish a data transmission scenario of UAV swarms to terrestrial users, including line-of-sight transmission and non-line-of-sight transmission and their corresponding probabilities;

[0007] (2) Model the UAV swarm task scheduling and channel allocation problem as a trajectory optimization and channel access game model based on potential game, where the game participants are the set of UAVs;

[0008] (3) Use a user scheduling and channel joint selection method based on best response to solve the trajectory optimization and channel access game model, so as to maximize the total utility of all terrestrial users served by each UAV, and complete the UAV swarm task scheduling and channel allocation.

[0009] Furthermore, step (1) is specifically as follows: In the described UAV mission execution scenario, there are N UAVs, and each UAV serves K ground users in sequence. There are M channels in the network. For UAV i, it transmits data directly above the served user. The channel model between the UAV and the user is divided into two types: line-of-sight transmission and non-line-of-sight transmission.

[0010] Furthermore, in step (1), the specific calculation method of the channel gain of the probability channel model is as follows:

[0011] For line-of-sight transmission, its probability is calculated as follows:

[0012] ;

[0013] where a and b are parameters describing the environment, is the elevation angle of the UAV, which is related to the UAV position and the ground user position; the elevation angle between UAV i and ground user n is specifically calculated as follows:

[0014] ;

[0015] where H is the UAV flight altitude, is the distance between UAV i and ground user n; define the coordinates of UAV i as , and the coordinates of ground user n as , specifically calculated as follows:

[0016] ;

[0017] For non-line-of-sight transmission, its probability is calculated as follows:

[0018] ;

[0019] For the defined probability channel model, the channel gain between UAV i and ground user n is:

[0020] ;

[0021] where, is the path loss exponent factor, is the additional attenuation coefficient due to non-line-of-sight transmission.

[0022] Furthermore, in step (2), the model formula is as follows:

[0023] ;

[0024] Among them, represents the set of UAVs, and N is the number of UAVs; is the set of action strategies of UAV i; is the utility function of UAV i.

[0025] Furthermore, in step (2), the utility function of UAV i is , and the specific definition is as follows:

[0026] For any UAV i, define as the channel selection of UAV i in K stages, as the service user scheduling selection of UAV i in K stages, as the sequential selection of UAV i to serve user k; based on this, the utility function of UAV i is defined as follows:

[0027] ;

[0028] Among them, k represents the serial number of the ground user served by UAV i, represents the transmission power of UAV i, represents the transmission power of UAV j, represents the channel gain between UAV j and the k-th ground user served by UAV i, and respectively represent the channel selections of UAV i and UAV j.

[0029] Furthermore, in step (3), the following steps are included:

[0030] (31) Initialization, the initial number of iterations is set to , and each UAV in the set of UAVs randomly selects a channel and the order of user services, and the initial strategy is defined as , among which, , represents the channel selection strategy vector of UAV i at the 0th iteration, , represents the user service order vector of UAV i at the 0th iteration;

[0031] (32) Fix the channel selections of all current UAVs , and use the best response algorithm to optimize the service user scheduling selections of the UAV swarm, and the optimization result is denoted as ;

[0032] (33) Fix the optimized service user scheduling selections of all current UAVs , and use the best response algorithm to optimize the channel selections of the UAV swarm, and the optimization result is denoted as ;

[0033] (34) Repeat steps (32) to (33) until the selection method converges; output the current utility of each UAV and the joint channel and user service scheduling selection results, and complete the UAV swarm task scheduling and channel allocation.

[0034] Further, in step (32), the optimal response algorithm is used to optimize the service user scheduling selection of the UAV swarm. The specific update criterion is as follows:

[0035] Randomly select a UAV i, and select the service user scheduling strategy corresponding to the maximum utility as the service user scheduling for the next iteration according to the following formula. The remaining UAVs repeat the previous service user scheduling:

[0036] ;

[0037] Wherein, is the service user scheduling selection of UAV i in the -th iteration, is the utility function of UAV i in the q -th iteration.

[0038] Further, in step (33), the optimal response algorithm is used to optimize the channel selection of the UAV swarm. The specific update criterion is as follows:

[0039] Randomly select a UAV i, and select the channel resource corresponding to the maximum utility as the channel selection for the next iteration according to the following formula. The remaining UAVs repeat the previous channel selection:

[0040] ;

[0041] Wherein, is the channel selection of UAV i in the -th iteration, is the utility function of UAV i in the q -th iteration.

[0042] An electronic device according to the present invention includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the computer program is loaded into the processor, it implements the method for UAV swarm task scheduling and channel allocation based on potential game described in any one of the above.

[0043] A storage medium according to the present invention stores a computer program, and when the computer program is executed by a processor, it implements the method for UAV swarm task scheduling and channel allocation based on potential game described in any one of the above.

[0044] Beneficial effects: Compared with the prior art, the present invention has the following remarkable advantages: By applying non - cooperative game, each unmanned aerial vehicle (UAV) has the ability of autonomous decision - making and makes the most beneficial decision for itself in a distributed manner, effectively reducing the computational complexity of the network. At the same time, through the proposed joint task scheduling and channel allocation framework, the dynamic interference between UAVs and devices can be effectively avoided, and the task execution utility of UAVs can be effectively improved. Brief Description of the Drawings

[0045] Figure 1 It is a schematic diagram of the data transmission scenario of the UAV swarm to ground users in the present invention.

[0046] Figure 2 It is a flowchart of the user scheduling and channel joint selection method based on best response in the present invention.

[0047] Figure 3 It is a simulation diagram of the user scheduling and channel joint selection method based on best response in the present invention. Detailed Embodiment

[0048] The technical solution of the present invention will be further described below with reference to the drawings.

[0049] Embodiment 1: As Figure 1 shown, the embodiment of the present invention provides a method for UAV swarm task scheduling and channel allocation based on potential game, including the following steps:

[0050] Step 1: In the UAV - assisted ground communication network, establish a data transmission scenario of the UAV swarm to ground users. In this scenario, a probabilistic channel model is considered, including two types: line - of - sight (LoS) transmission and non - line - of - sight (NLoS) transmission and their corresponding probabilities. Specifically, in the UAV task execution scenario, there are N UAVs, each UAV serves K ground users in sequence, and there are M channels in the network;

[0051] For LoS transmission, the probability is calculated as follows:

[0052] ;

[0053] where a and b are parameters describing the environment, is the elevation angle of the UAV, which is related to the UAV position and the ground user position; the elevation angle between UAV i and ground user n is calculated as follows:

[0054] ;

[0055] where H is the UAV flight altitude, is the distance between UAV i and ground user n; define the coordinates of UAV i as , the coordinates of the ground user n are , and the specific calculation method is as follows:

[0056] ;

[0057] For non-line-of-sight transmission, its probability is calculated as follows:

[0058] ;

[0059] For the defined probability channel model, the channel gain between the UAV i and the ground user n is:

[0060] ;

[0061] Among them, is the path loss exponent factor, is the additional attenuation coefficient due to non-line-of-sight transmission.

[0062] Step 2: Model the UAV swarm task scheduling and channel allocation problem as a trajectory optimization and channel access game model based on potential game, where the game participants are the UAV set, specifically as follows:

[0063] Model the cooperative offloading problem as a channel access game model based on task offloading, and this game model is defined as:

[0064] ;

[0065] Among them, represents the UAV set, and N is the number of UAVs; is the action strategy set of the UAV i is the utility function of the UAV i. For any UAV i, define as the channel selection of the UAV i in K stages, as the service user scheduling selection of the UAV i in K stages, as the order selection of the UAV i to serve the user k; based on this, the utility function of the UAV i is defined as follows:

[0066] ;

[0067] Among them, k represents the serial number of the ground user served by the UAV i, represents the transmission power of the UAV i, represents the transmission power of the UAV j, represents the channel gain between the UAV j and the k-th ground user served by the UAV i, and respectively represent the channel selection of UAV i and UAV j.

[0068] Step 3: Propose a user scheduling and channel joint selection method based on the best response to solve the game, so as to maximize the total utility of all ground users served by each UAV. The user scheduling and channel joint selection method based on the best response is as follows:

[0069] (31) Initialization. Set the initial number of iterations to , and each UAV in the UAV set randomly selects a channel and the user service order. Define the initial strategy as , where , represents the channel selection strategy vector of UAV i at the 0th iteration, , represents the user service order vector of UAV i at the 0th iteration;

[0070] (32) Fix the channel selection of all current UAVs , and use the best response algorithm to optimize the service user scheduling selection of the UAV swarm. The optimization result is denoted as ; The specific update criterion is as follows:

[0071] Randomly select a UAV i, and select the service user scheduling strategy corresponding to the maximum utility as the service user scheduling for the next iteration according to the following formula. The remaining UAVs repeat the previous service user scheduling:

[0072] ;

[0073] where is the service user scheduling selection of UAV i in the th iteration, is the utility function of UAV i in the qth iteration.

[0074] (33) Fix the optimized service user scheduling selection of all current UAVs , and use the best response algorithm to optimize the channel selection of the UAV swarm. The optimization result is denoted as ; The specific update criterion is as follows:

[0075] Randomly select a UAV i, and select the channel resource corresponding to the maximum utility as the channel selection for the next iteration according to the following formula. The remaining UAVs repeat the previous channel selection:

[0076] ;

[0077] where is the Channel selection of UAV i in the q-th iteration is the utility function of UAV i in the q-th iteration.

[0078] (34) Repeat steps (32) to (33) until this selection method converges; output the utilities of current UAVs and the joint channel and user service scheduling selection results to complete the task scheduling and channel allocation of the UAV swarm.

[0079] Embodiment 2: The present invention provides a method for task scheduling and channel allocation of a UAV swarm based on potential game, including the following steps:

[0080] S1: Consider a scenario of data transmission from a UAV swarm to ground users, as Figure 1 shown, including N UAVs, each UAV serving K ground users in sequence, and 2 channels. Each UAV serves 3 ground users in a certain order.

[0081] For line-of-sight transmission, its probability is calculated as follows:

[0082] ;

[0083] where a and b are parameters describing the environment, taking and . is the elevation angle of the UAV, which is related to the UAV position and the ground user position; the elevation angle between UAV i and ground user n is calculated as follows:

[0084] ;

[0085] where H is the flight altitude of the UAV, , is the distance between UAV i and ground user n; define the coordinates of UAV i as , and the coordinates of ground user n as , calculated as follows:

[0086] ;

[0087] For non-line-of-sight transmission, its probability is calculated as follows:

[0088] ;

[0089] For the defined probability channel model, the channel gain between UAV i and ground user n is:

[0090] ;

[0091] Among them, is the path loss exponent factor, , is the additional attenuation coefficient due to non-line-of-sight transmission, .

[0092] S2: Model the UAV swarm task scheduling and channel allocation problem as a trajectory optimization and channel access game model based on potential game, where the game participants are the UAV set, specifically as follows:

[0093] Model the cooperative offloading problem as a channel access game model based on task offloading, and this game model is defined as:

[0094] ;

[0095] Among them, represents the UAV set, and N is the number of UAVs; is the action strategy set of UAV i is the utility function of UAV i. For any UAV i, define as the channel selection of UAV i in K stages, as the service user scheduling selection of UAV i in K stages, as the sequential selection of UAV i serving user k; based on this, the utility function of UAV i is defined as follows:

[0096] ;

[0097] Among them, k represents the serial number of the ground user served by UAV i, represents the transmission power of UAV i, represents the transmission power of UAV j, take . represents the channel gain between UAV j and the kth ground user served by UAV i, and respectively represent the channel selections of UAV i and UAV j.

[0098] S3: Propose a user scheduling and channel joint selection method based on the best response to solve the game, so as to maximize the total utility of all ground users served by each UAV. Among them, the user scheduling and channel joint selection method based on the best response is as Figure 2 shown, specifically as follows:

[0099] (S31) Initialization, set the initial iteration number to , and each UAV in the UAV set randomly selects a channel and the user service order, and define the initial strategy as , where , represents the channel selection strategy vector of UAV i at the 0th iteration, , represents the user service order vector of UAV i at the 0th iteration;

[0100] (S32) Fix the channel selection of all current UAVs , and use the best response algorithm to optimize the service user scheduling selection of the UAV swarm. The optimization result is denoted as ; The specific update criterion is as follows:

[0101] Randomly select a UAV i, and select the service user scheduling strategy corresponding to the maximum utility as the service user scheduling for the next iteration according to the following formula. The remaining UAVs repeat the previous service user scheduling:

[0102] ;

[0103] where is the service user scheduling selection of UAV i in the th iteration, is the utility function of UAV i in the qth iteration.

[0104] (S33) Fix the optimized service user scheduling selection of all current UAVs , and use the best response algorithm to optimize the channel selection of the UAV swarm. The optimization result is denoted as ; The specific update criterion is as follows:

[0105] Randomly select a UAV i, and select the channel resource corresponding to the maximum utility as the channel selection for the next iteration according to the following formula. The remaining UAVs repeat the previous channel selection:

[0106] ;

[0107] where is the channel selection of UAV i in the th iteration, is the utility function of UAV i in the qth iteration.

[0108] (S34) Repeat steps (S32) to (S33) until this selection method converges; output the utility of each current UAV and the combined channel and user service scheduling selection result to complete the task scheduling and channel allocation of the UAV swarm.

[0109] Consider a UAV mission execution scenario where the number of UAVs changes from 2 to 6 and the number of ground users changes from 6 to 18. The number of channels in the fixed network is 2. Figure 3This is the result of the experimental run.

[0110] As Figure 3 shown, it is an experimental simulation diagram of the relationship between the overall network task utility and the number of UAVs (number of users) under the application of the proposed user scheduling and channel joint selection method based on optimal response, the user scheduling optimization algorithm, the channel selection optimization algorithm, and the random selection algorithm. It can be seen that the global utility of the method designed by the present invention is significantly higher than that of the other three methods. Moreover, as the number of UAVs (number of users) increases, the utility gradually decreases, and the performance gap with random selection becomes larger and larger.

[0111] Example 3: The embodiment of the present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the computer program is loaded into the processor, it implements any one of the methods for UAV swarm task scheduling and channel allocation based on potential game.

[0112] Example 4: The embodiment of the present invention also provides a storage medium storing a computer program, and when the computer program is executed by a processor, it implements any one of the methods for UAV swarm task scheduling and channel allocation based on potential game.

Claims

1. A method for task scheduling and channel allocation of drone swarms based on potential energy game, characterized in that: The following steps are involved: (1) In the UAV swarm-assisted ground communication network, the data transmission scenarios of the UAV swarm to the ground users are established, including line-of-sight transmission and non-line-of-sight transmission and their corresponding probabilities. The specific calculation method of the channel gain of the probabilistic channel model is as follows: For line-of-sight transmission, the probability The calculation method is as follows: ； Among them, a and b are parameters describing the environment, is the elevation angle of the drone, which is related to the position of the drone and the position of the ground user; the elevation angle between drone i and ground user n , the specific calculation method is as follows: ； Among them, H is the flight altitude of the UAV, is the distance between UAV i and ground user n; the coordinates of UAV i are defined as , the coordinates of ground user n are , the specific calculation method is as follows: ； For non-line-of-sight transmission, the probability The calculation method is as follows: ； The probabilistic channel model defined by for: ； in, is the path loss exponential factor, is the additional attenuation coefficient due to non-line-of-sight transmission; (2) The UAV swarm task scheduling and channel allocation problem is modeled as a trajectory optimization and channel access game model based on potential energy game, where the game participants are a collection of UAVs; (3) Solving the track optimization and channel access game model using the optimal response-based user scheduling and channel joint selection method, so as to maximize the total utility of all ground users served by each drone, and complete the drone swarm task scheduling and channel allocation; including the following steps: (31) Initialization, the initial number of iterations is set to , each drone in the drone set randomly selects the channel and user service order, and defines the initial strategy as ,in, , represents the channel selection strategy vector of UAV i at iteration 0, , represents the user service order vector of drone i at iteration 0; (32) Fixed the channel selection for all current drones , the optimal response algorithm is used to optimize the service user scheduling selection of the drone group, and the optimization result is recorded as ; (33) Fixed the optimized service user scheduling selection for all current drones , the optimal response algorithm is used to optimize the channel selection of the drone group, and the optimization result is recorded as ; (34) Repeat steps (32) to (33) until the selection method converges; output the utility of each current UAV and the joint channel and user service scheduling selection results to complete the UAV group task scheduling and channel allocation.

2. The method for UAV swarm task scheduling and channel allocation based on potential game according to claim 1 is characterized in that: Step (1) is as follows: Assume that the UAV mission execution scenario includes N UAVs, each of which serves K ground users in sequence, and there are M channels in the network; for UAV i, it transmits data directly above the user it serves; the channel model between the UAV and the user is divided into line-of-sight transmission and non-line-of-sight transmission.

3. The method for UAV swarm task scheduling and channel allocation based on potential energy game according to claim 1 is characterized in that: In step (2), the model formula is as follows: ； in, represents the set of drones, N is the number of drones; is the action strategy set of drone i is the utility function of UAV i.

4. The method for UAV swarm task scheduling and channel allocation based on potential game according to claim 3 is characterized in that: In step (2), the utility function of drone i is , which is defined as follows: For any drone i, define is the channel selection of UAV i in K stages, For the scheduling selection of service users for UAV i in K stages, The sequential selection of drone i to serve user k; based on this, the utility function of drone i is defined as follows: ； Where k represents the ground user number served by UAV i, represents the transmission power of drone i, represents the transmission power of UAV j, represents the channel gain between UAV j and the kth ground user served by UAV i, and They represent the channel selection of UAV i and UAV j respectively.

5. The method for UAV swarm task scheduling and channel allocation based on potential game according to claim 1 is characterized in that: In step (32), the optimal response algorithm is used to optimize the service user scheduling selection of the drone group. The specific update criteria are as follows: Randomly select a drone i, and select the service user scheduling strategy corresponding to the maximum utility as the service user scheduling for the next iteration according to the following formula. The remaining drones repeat the previous service user scheduling: ； in, It is The service user scheduling selection of drone i in the iteration, is the utility function of drone i in the qth iteration.

6. The method for UAV swarm task scheduling and channel allocation based on potential energy game according to claim 1 is characterized in that: In step (33), the optimal response algorithm is used to optimize the channel selection of the drone group. The specific update criteria are as follows: Randomly select a drone i, and select the channel resource corresponding to the maximum utility as the channel selection for the next iteration according to the following formula. The remaining drones repeat the previous channel selection: ； in, It is The channel selection of drone i in the iteration, is the utility function of drone i in the qth iteration.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is loaded into the processor, a method for unmanned aerial vehicle swarm task scheduling and channel allocation based on potential energy game according to any one of claims 1 to 6 is implemented.

8. A storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, a method for unmanned aerial vehicle swarm task scheduling and channel allocation based on potential energy game according to any one of claims 1 to 6 is implemented.

Images