Short-Packet Communications for UAV-Based NOMA Systems Under Imperfect CSI and SIC

This paper studies the performance of unmanned aerial vehicle (UAV)-aided short-packet non-orthogonal multiple access (NOMA) systems under imperfect channel state information (ipCSI) and successive interference cancellation (ipSIC). Closed-form expressions for the average block-error rate (BLER), throughput, goodput, latency, reliability, and age of information (AoI) are derived. To gain useful insights, an asymptotic BLER analysis is carried out at high signal-to-noise ratio (SNR) regime, from which users' diversity order and the impact of the UAV's trajectory are deduced. Three optimization problems are formulated in order to: 1) minimize the overall BLER of the user with high quality-of-service by optimizing the power allocation (PA) coefficients; 2) maximize the system throughput by jointly optimizing PA and coding rate (CR) coefficients; 3) minimize AoI by optimizing block-length transmission per-hop transmission. Numerical results show that, at high SNR regime, users' performance is not influenced by UAVs' radius and horizontal flight angle when ipCSI occurs. The proposed system achieves better throughput than the baseline schemes and can be further enhanced by jointly optimizing PA and CR. To achieve the minimum AoI, the source and UAV should transmit a moderate block-length for low transmit power, while for moderate and high transmit power, 100 is sufficient.