A joint optimization method for UAV scheduling and trajectory planning in non-orthogonal multiple access networks

Abstract

The present invention proposes a joint optimization method for UAV scheduling and trajectory planning in a non-orthogonal multiple access network. The present invention is based on a UAV-assisted non-orthogonal multiple access wireless network system with multiple ground users, in which the main task of the UAV is to schedule ground users to provide services and meet the service quality requirements of users. The present invention designs a reasonable decoding order for multiple users in a non-orthogonal multiple access wireless network according to the channel gain of the users, and forms a scheduling method to enable the UAV to select appropriate ground users to provide services at different times. Under the constraints of meeting the maximum transmission power, the service quality requirements of each user and the maximum flight speed of the UAV, multiple track points are obtained and connected into an optimized trajectory. The method proposed in the present invention can design an adaptive scheduling method and an optimal flight trajectory for the UAV providing services in a UAV-assisted non-orthogonal multiple access wireless network system with multiple ground users, on the premise of ensuring the service quality requirements of the scheduled ground users, so as to improve the throughput of the entire system.

Description

Technical Field

[0001] The present invention relates to user scheduling and trajectory optimization when a drone is used to serve ground users in a drone-assisted non-orthogonal multiple access wireless network. The present invention is mainly used to design an optimal scheduling scheme and plan an optimal flight trajectory for the drone when the drone performs a mission, under the premise of ensuring the minimum data rate requirements of all ground users and satisfying the maximum transmission power of the drone. Background Art

[0002] Unmanned aerial vehicles (UAVs) have attracted widespread attention in the past few decades in various applications such as surveillance and detection, aerial imaging, cargo transportation, search and rescue, etc. They can not only carry base stations to relieve temporary communication pressure for emergency rescue and disaster relief, but also combine with existing cellular networks to provide services to ground users. As one of the various applications of UAVs, UAV communication has become more and more prominent with the development of auxiliary communication derivative solutions, as it can be equipped with communication devices and large-capacity batteries to meet the quality of service (QoS) requirements for on-demand deployment. Based on these advantages, many commercial companies have begun to focus on UAVs and related fields for a series of research. Although UAVs add new degrees of freedom to the network, they also bring various new challenges in communication, such as user scheduling strategies, trajectory planning, and transmission technology. Summary of the invention

[0003] Purpose of the invention: Due to the sparse distribution of ground users and the different service quality requirements of each user, in a non-orthogonal multiple access wireless network, the number of users that can be scheduled by a drone at different locations during flight is also different. Therefore, how to optimize the scheduling scheme and trajectory so that the drone can adaptively schedule users and maximize the throughput of the entire system has become a major technical issue.

[0004] To solve this technical problem, the present invention proposes a joint optimization method for UAV scheduling and trajectory planning in a non-orthogonal multiple access network, which maximizes the total throughput of the system by optimizing the scheduling scheme and the corresponding trajectory.

[0005] Technical solution: To achieve the above technical effects, the technical solution proposed by the present invention is:

[0006] A joint optimization method for unmanned aerial vehicle scheduling and trajectory planning in a non-orthogonal multiple access network comprises the following steps:

[0007] (1) Constructing the system model (assuming that the drone flies in a three-dimensional environment, the drone starts from the origin and circles once and returns to the origin within a time period T. The time period T is divided into N time slots. Assuming that the drone flies at a fixed altitude H, the horizontal coordinate of the drone in the nth time slot is q n =[x n,y n ] T , further, there are M ground users in total, and the coordinates of the mth ground user are represented by u m =[x m ,y m ] T ):

[0008] 1) Assume that the maximum flight speed of the drone is v max , the drone trajectory satisfies the following conditions:

[0009] q 1 =q N (1)

[0010] ||q n+1 -q n ||≤S max , n=1,...N-1 (2)

[0011] Among them, S max =v max (T / N) indicates the maximum flight distance of the drone in a time slot.

[0012] 2) The distance between the UAV and the ground user m in the nth time slot is:

[0013]

[0014] 3) Assuming that the channel gain follows the free space path loss model, and the communication link from the drone to the ground user is regarded as line-of-sight wireless transmission, the channel quality depends on the distance, then the channel power gain can be expressed as:

[0015]

[0016] 4) Within the duration T, the drone can match different numbers of users in different time slots. Non-orthogonal multiple access technology can cause signal interference to users sharing the same channel. SIC decoding technology can eliminate interference one by one according to the channel quality order. Users with poor channel conditions will be interfered by users with good channel conditions. The signal-to-noise ratio of user m in the nth time slot is:

[0017]

[0018] in, N represents the interference caused to user m by other users who share the same channel with user m, 0 represents the additive white Gaussian noise power, p n，m Indicates the transmission power of user m in the nth time slot.

[0019] 5) The power allocated to user m in the nth time slot is:

[0020]

[0021] 6) The data rate of user m in the nth time slot is:

[0022] r n，m =log 2 (1+SINR n，m (p n ,q n )) (7)

[0023] 7) During the duration T, the total data rate of the drone is:

[0024]

[0025] 8) Taking into account the service quality requirements of different users (the data rate of each scheduled user must meet the minimum rate requirement), the maximum transmission power that the drone can provide (the upper bound of the sum of the powers allocated to all users scheduled in a time slot), the trajectory constraints of the drone (the upper bound of the maximum flight distance of the drone in each time slot is the distance that the drone always flies at the maximum flight speed, and the starting point and end point of the drone are the same) and the scheduling status of the user in each time slot, the maximum system throughput in the drone-based non-orthogonal multiple access wireless network is:

[0026]

[0027] (2) We propose an alternating iterative algorithm to solve the rate maximization problem in non-orthogonal multiple access wireless networks based on drones. For the user scheduling problem, we design a user admission scheme. For the drone trajectory planning problem, we use the continuous convex approximation method to transform the non-convex problem into a convex problem and solve it using CVX. The drone trajectory and user scheduling are iterated alternately until convergence.

[0028] 1) For the user scheduling problem, a user admission scheme is proposed to calculate the maximum number of users to meet the target data rate of each user. The formulation of this sub-optimization problem is as follows:

[0029]

[0030] Among them, the drone forms a preference list based on the channel gain of ground users. The users at the back of the list will be interfered by the users at the front of the list and are scheduled. Each scheduled user will be allocated the minimum power that can meet its needs. The drone can adaptively schedule ground users according to their positions in different time slots during the flight.

[0031] 2) For the trajectory optimization problem, the scheduling results obtained in the user admission scheme are used to transform the rate maximization problem in the wireless network based on drones into an optimization problem related only to the trajectory of drones:

[0032]

[0033] Where R(Q) represents the sum rate of given user scheduling and UAV transmit power allocation.

[0034] Note that due to the existence of trajectory variables in the objective function, problem (11) is not a convex optimization problem. We use the continuous convex approximation method to transform the non-convex problem into a convex optimization problem and finally express it as:

[0035]

[0036] After trajectory optimization, users whose target data rates are not met will be eliminated. Problem (12) is a convex optimization problem, and the optimal value is obtained by CVX solver.

[0037] Based on the user access scheme and drone trajectory optimization, we use an iterative algorithm to alternately iterate two variable blocks. When optimizing one variable block, we ensure that the other one is a fixed value. The result of the previous iteration will be used in the next one until convergence to the optimal value.

[0038] Beneficial effects: The present invention can improve the overall system throughput and the average response time of users during UAV communication by optimizing the scheduling scheme and flight trajectory, and improve energy efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Schematic diagram of a non-orthogonal multiple access wireless network link between a UAV and multiple ground users.

[0040] Figure 2 is a graphical illustration of an iterative algorithm.

[0041] Figure 3 It is a graph of ground user results for UAV scheduling based on the user access scheme.

[0042] Figure 4 This is the trajectory diagram of the drone serving 8 ground users before and after optimization. DETAILED DESCRIPTION

[0043] The present invention will be further described below in conjunction with the accompanying drawings.

[0044] Figure 1Schematic diagram of a non-orthogonal multiple access wireless network link between a UAV and multiple ground users. The UAV acts as an aerial base station to provide services to ground users at a fixed altitude H and no more than v max The UAV needs to return to the origin after a time period T to restart the next service cycle. m =[x m ,y m ] T is represented by the coordinates of ground user m. In addition, q n =[x n ,y n ] T Represents the horizontal coordinate of the drone at the nth time slot.

[0045] Figure 2 is a graphic illustration of the iterative algorithm. Our objective function is to maximize the throughput that the entire system can obtain during the entire cycle. The throughput of the system is maximized by comprehensively considering the different service quality requirements of ground users, the maximum transmission power of the communicating UAV, the flight trajectory of the UAV, and the scheduling status of the ground users in each time slot. We decompose the problem into two parts: user scheduling and trajectory planning. First, we use the UAV trajectory obtained in the last iteration to solve the user scheduling and power allocation, and then use the known user scheduling results to obtain the optimized trajectory in a similar way. Alternately iterate these two variable blocks until convergence to the optimal value.

[0046] Figure 3 This is a diagram of ground user results of drone scheduling based on the user admission scheme. The left and right are user scheduling result diagrams after trajectory optimization adjustment. All users are randomly distributed in a two-dimensional area of ​​1×1 square kilometers. The users selected by the drone for service in a certain time slot are adjusted in the process of alternating iterations of the user admission method and drone trajectory optimization. Users who do not meet their minimum data rate requirements are deleted from the scheduled user set to ensure the service quality of each scheduled user and to achieve a trade-off between the number of users and system throughput. The missing connection line on the right is the user who has changed from a scheduled state to an unscheduled state after adjustment. When the number of ground users increases, the user distribution becomes more dense, and the drone can provide services to more users in each time slot, thereby reducing the user waiting time for service and improving the throughput of the entire system.

[0047] Figure 4The trajectory diagrams before and after optimization for drones serving 8 ground users. Ground users have different service quality requirements. The ring marked by triangles is the initial drone trajectory, and the ring marked by circles is the optimized drone trajectory. The changes in the drone flight trajectory can be roughly seen. The drone dynamically adjusts its position to better provide services based on the distribution of ground users and the data rate requirements of each user. It can be seen that the optimized trajectory is more biased towards users who deviate from the initial trajectory, so that these initially deviated users can get better service quality.

[0048] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A joint optimization method for UAV scheduling and trajectory planning in a non-orthogonal multiple access network, characterized in that: Includes steps: (1) Assume that the drone flies in a three-dimensional environment. The drone starts from the origin and makes a circle and returns to the origin within a time period T. The time period T is divided into N time slots. Assume that the drone flies at a fixed altitude H. The horizontal coordinate of the drone in the nth time slot is q n =[x n ,y n ] T , further, there are M ground users in total, and the coordinates of the mth ground user are represented by u m =[x m ,y m ] T , build the system model: 1) Assume that the maximum flight speed of the drone is v max , the drone trajectory satisfies the following conditions: q1=q N (1) ||q n+1 -q n ||≤S max ,n=1,…N-1(2) Among them, S max =v max (T / N) indicates the maximum flight distance of the drone in a time slot; 2) The distance between the UAV and the ground user m in the nth time slot is: 3) Assuming that the channel gain follows the free space path loss model, and the communication link from the drone to the ground user is regarded as line-of-sight wireless transmission, the channel quality depends on the distance, then the channel power gain can be expressed as: 4) Within the duration T, the drone can match different numbers of users in different time slots. Non-orthogonal multiple access technology can cause signal interference to users sharing the same channel. SIC decoding technology can eliminate interference one by one according to the channel quality order. Users with poor channel conditions will be interfered by users with good channel conditions. The signal-to-noise ratio of user m in the nth time slot is: in, N represents the interference caused to user m by other users who share the same channel with user m, o represents the additive white Gaussian noise power, p n，m Indicates the transmission power of user m in the nth time slot; 5) The power allocated to user m in the nth time slot is: 6) The data rate of user m in the nth time slot is: r n，m =log2(1+SINR n，m (p n ,q n )) (7) 7) During the duration T, the total data rate of the drone is: 8) Comprehensively consider the service quality requirements of different users, that is, the data rate of each scheduled user must meet the minimum rate requirement, the maximum transmission power that the drone can provide, that is, the upper bound of the sum of the powers allocated to all users scheduled in a time slot, the trajectory constraint of the drone, that is, the upper bound of the maximum flight distance of the drone in each time slot is the distance that the drone always flies at the maximum flight speed, and the starting and ending points of the drone are consistent, as well as the scheduling status of the user in each time slot. In the non-orthogonal multiple access wireless network based on drones, the maximum system throughput is: (2) An alternating iterative algorithm is proposed to solve the rate maximization problem in non-orthogonal multiple access wireless networks based on drones. A user admission scheme is designed for the user scheduling problem. For the drone trajectory planning problem, a continuous convex approximation method is used to transform the non-convex problem into a convex problem and solve it using CVX. The drone trajectory and user scheduling are iterated alternately until convergence. 1) For the user scheduling problem, a user admission scheme is proposed to calculate the maximum number of users to meet the target data rate of each user. The user admission scheme is described as follows: Among them, the drone forms a preference list based on the channel gain of the ground users. The users at the back of the list will be interfered by the users at the front of the list and are scheduled. Each scheduled user will be allocated the minimum power that can meet its needs. The drone can adaptively schedule the ground users according to their positions in different time slots during the flight. 2) For the trajectory optimization problem, using the scheduling results obtained from the previous user admission scheme, the rate maximization problem in the wireless network based on drones is transformed into an optimization problem related only to the drone trajectory: Where R(Q) represents the sum rate of given user scheduling and UAV transmit power allocation; Due to the existence of trajectory variables in the objective function, problem (11) is not a convex optimization problem. The continuous convex approximation method is used to transform the non-convex problem into a convex optimization problem and finally expressed as: After trajectory optimization, users whose target data rates are not met will be eliminated; Problem (12) is a convex optimization problem, and the optimal value is obtained by using the CVX solver; 3) Based on the user access scheme and drone trajectory optimization, an iterative algorithm is used to alternately iterate two variable blocks. When optimizing one of the variable blocks, the other is ensured to be a fixed value. The result of the previous iteration will be used in the next one until convergence to the optimal value.

2. The method for joint optimization of UAV scheduling and trajectory planning in a non-orthogonal multiple access network according to claim 1, characterized in that: In a UAV-assisted non-orthogonal multiple access wireless network with multiple ground users, an optimal scheduling scheme and the corresponding optimal flight trajectory are designed for the UAVs providing services with the goal of system throughput, while ensuring the service quality of all ground users and satisfying the constraint of the maximum transmission power of the UAVs.

Images