Deep Learning Based Communication for Power-Efficient Satellite Systems

This project aims at deriving new communication techniques based on deep learning to improve the overall performance of satellite communication systems. The special focus is on the reduction of peak-to-average power ratio (PAPR) and nonlinear distortion of multi-subcarrier signals to gain high power efficiency in the emerging large-scale low Earth orbit (LEO) satellite constellations. Autoencoder architectures using deep neural networks (DNNs) will be adopted as the main solution to achieve the goal.