3D UAV Placement and Resource Allocation in Software Defined Cellular Networks

Abstract

A promising solution to satisfying the demand for flexible and elastic networks in an unexpected or temporary event is to constitute drone-cells, which are formed by quick deployment of unmanned aerial vehicles (UAVs). The multidimensional ground-air resources in the drone-assisted networks make resource allocation a critical issue in future cellular networks. A UAV placement and resource allocation algorithm is proposed that utilizes the global view and the elastic characteristics of software defined cellular networks (SDCNs). A utility maximization problem is formulated to study the reasonable three-dimensional (3D) UAV placement and optimize the tradeoff between the associated number of users and the transmission power of UAVs, which also keeps the associated users’ QoS requirement higher than the threshold. After mathematical manipulations, the intractable utility maximization problem is converted into a two-phase algorithm involving optimal drone-cell altitude-to-radius ratio, and the optimal resource allocation. Simulation results indicate the growth of the average achievable data rate of users and the network tradeoff utility of the proposed algorithm, compared with the fixed transmission power for optimal UAV placement and two-dimensional (2D) UAV placement for optimal resource allocation schemes. By deploying the new designs, the maximum data rate gain can reach up to 39.79%.