Formation Configuration Optimization for UAV Swarm Enabled Multi-User Communications

Abstract

This paper investigates a UAV swarm enabled multi-user downlink communication system, where multiple singleantenna UAVs cooperate as a virtual antenna array (VAA) to serve ground users. Inspired by flexible antenna techniques, the unmanned aerial vehicle (UAV) swarm is modeled as a rigid formation whose reference point and heading angle can be jointly optimized to reconfigure the air-to-ground (A2G) channels. We formulate a max-min SINR problem by jointly optimizing the UAV formation configuration and transmit beamforming vectors. To solve the resulting non-convex problem, a hierarchical algorithm is proposed, where adaptive particle swarm optimization (APSO) is used for formation configuration optimization, while regularized zero-forcing (RZF) beamforming and bisection-based power allocation are adopted for low-complexity fitness evaluation. Numerical results show that the proposed scheme improves the worst-user SINR compared with benchmark schemes.