| With the popularization of UAV(Unmanned Aerial Vehicle)technology,the development potential of UAV in many fields has been gradually explored,especially in the field of wireless communication.The UAV low-altitude platform can not only make up for the shortcomings of the traditional ground communication system,but also form an air communication system that can be flexibly deployed to the demand scenario to provide high-quality continuous services to the ground request users.By designing a reasonable system structure,the coverage capability of the entire communication system can be effectively optimized.By optimizing the deployment location of the system,the service quality and resource utilization of the system can be effectively improved.In this thesis,the optimal deployment of UAV low-altitude platform communication system is studied.The deployment system adopts the regional division strategy and the step-by-step allocation method,and the optimal deployment of the whole system is carried out through the optimization model and algorithm in this thesis.Firstly,this thesis introduces the performance parameters of different types of UAV and airborne equipment,selects appropriate equipment to construct the UAV low altitude platform in this thesis,and applies the actual performance parameters to the simulation experiment in this thesis.Aiming at the actual user distribution problem,the simulation scene is improved.By combining Poisson cluster process and uniform distribution,the concept of hot spot area is proposed,which enhances the authenticity and complexity of the initial simulation scene.Then,the DDP(Dynamic Density Peak)algorithm is proposed to solve the problem of region division in user distribution scenarios.By limiting the area coverage and the maximum number of service users,the region division in user distribution scenarios is carried out.DDP algorithm can automatically select the center point of the initial area by calculating the local density of the user position coordinates.According to the location of the center point of the region,the nearest principle is used to determine the ownership of the user,so that each user in the scene has a belonging area to ensure full coverage of the user in the current scene.DDP algorithm adopts a dynamic adjustment strategy,which can balance the number of users in each partition area and obtain the partition result of relatively balanced system load.The experiment proves the advantages of DDP algorithm in dealing with the problem of regional division.Finally,this thesis proposes the deployment problem of joint optimization of system throughput and communication processing power consumption.Considering the limited spectrum resources and the limited number of sub-channels caused by the performance limitation of airborne communication equipment,this thesis adopts the distributed allocation of platform channel and power to reduce the allocation overhead of airborne equipment.In the system,frequency division multiplexing is used between each platform region to reduce the mutual interference between the same frequency regions during platform deployment,and the performance differences of system deployment under different multiplexing factors are compared.Through the comparison and verification of simulation experiments,under different user scenarios and deployment schemes,the NSGA2(Non-dominated Sorting Genetic Algorithm II)algorithm combined with power injection allocation can effectively solve the joint optimal deployment problem of system throughput and communication processing power consumption,and give the optimal solution set of Pareto domain under different multiplexing factors.According to the actual demand,the optimal solution of the solution set can be selected for deployment. |
