Joint resource and trajectory optimization for video streaming in UAV-based emergency indoor-outdoor communication

Abstract

Due to its ability for flexible placement, the Unmanned Aerial Vehicle (UAV) has been widely utilized as an aerial relay to transmit the video streaming data, which is particularly useful in emergency communication networks that lack the communication infrastructure. In this paper, we examine the combined resource optimization and trajectory planning for video services using dynamic adaptive streaming over HTTP (DASH) in the uplink transmission of UAV-based emergency indoor-outdoor communication networks. In detail, a rotary-wing UAV works as an aerial relay, forwarding emergency video collected by indoor users to a remote Base Station (BS). To ensure that the UAV can forward the emergency video data from disaster area with stable video quality, we formulate an optimization problem of maximizing the uplink throughput and video streaming utility for all indoor users, by jointly optimizing the 3D flight trajectory of UAV, playback rate of video, communication time as well as bandwidth allocation, subject to the UAV trajectory constraints, total available bandwidth limitation, video quality variation constraints, as well as information-causality constraints for both UAV relaying and video playing. To tackle the formulated non-convex problem, we present a joint UAV trajectory, bandwidth allocation and video utility (JTBU) algorithm. Specifically, we first decouple the original problem into two sub-issues: 3D UAV trajectory optimization subproblem and resource allocation optimization subproblem. Then the JTBU algorithm alternately optimizes the two subproblems until the convergency is reached. Finally, the numerical results confirm that the proposed algorithm had a higher video streaming utility and system throughput than the baseline methods.