Channel Estimation for UAV Communication Systems Using Deep Neural Networks

• DOI: 10.3390/drones6110326
• Publication Date: 2022-10-30T03:45:00Z
• Authors: Ahmed Al-Gburi; Osamah Abdullah; Akram Y. Sarhan; Hayder Al-Hraishawi
• Keywords: Deep Neural Networks

Unmanned aerial vehicle (UAV) is steadily growing as a promising technology for next-generation communication systems due to their appealing features such wide coverage with high altitude, on-demand low-cost deployment, and fast responses. UAV communications are fundamentally different from the conventional terrestrial satellite owing mobility unique channel characteristics of air-ground links. However, obtaining effective state information (CSI) challenging because dynamic propagation environment variable transmission delay. In this paper, deep learning (DL)-based CSI prediction framework proposed address aging problem by extracting most discriminative wireless signals. Specifically, we develop procedure multiple Gaussian Bernoulli restricted Boltzmann machines (GBRBM) dimension reduction pre-training utilization incorporated an autoencoder-based neural networks (DNNs). To evaluate approach, real data measurements communicating base-stations within commercial cellular network obtained used training validation. Numerical results demonstrate that method accurate in acquisition various flying scenarios outperforms DNNs.