Machine Learning for CSI Recreation Based on Prior Knowledge
This work addresses CSI reporting overhead for mobile wireless systems, but it is incremental as it builds on existing ML methods like UNNs and cGANs.
The authors tackled the problem of channel state information (CSI) recreation in mobile wireless communications by combining untrained neural networks and conditional generative adversarial networks to model channels based on prior knowledge, achieving robust performance in line-of-sight conditions with low overhead reporting.
Knowledge of channel state information (CSI) is fundamental to many functionalities within the mobile wireless communications systems. With the advance of machine learning (ML) and digital maps, i.e., digital twins, we have a big opportunity to learn the propagation environment and design novel methods to derive and report CSI. In this work, we propose to combine untrained neural networks (UNNs) and conditional generative adversarial networks (cGANs) for MIMO channel recreation based on prior knowledge. The UNNs learn the prior-CSI for some locations which are used to build the input to a cGAN. Based on the prior-CSIs, their locations and the location of the desired channel, the cGAN is trained to output the channel expected at the desired location. This combined approach can be used for low overhead CSI reporting as, after training, we only need to report the desired location. Our results show that our method is successful in modelling the wireless channel and robust to location quantization errors in line of sight conditions.