RIS application at EDGE nodes

The dataset contains Channel State Information of RIS-assisted wireless environments, together with configuration indexes as labels. Specifically, it provides realizations of the channels 'h' and 'g' from a single antenna transmitter to the RIS and the RIS to the single antenna receiver, respectively. Each realization is labeled by the optimal RIS configuration that maximizes the receiver’s rate. 

The target group for this dataset can typically be engineers and researchers from wireless communications, signal processing, and machine learning. The rate-maximization by tuning the reflecting elements of RISs in a RIS setting is a combinatorial optimization problem, and therefore to compute an approximate configuration for the reflecting elements in an efficient manner is of crucial importance in practice. In this respect, the above dataset can be used to train a neural network for efficiently computing a reasonable configuration for reflecting elements. The dataset is not based on real channel measurements. It is worth pointing out that the use is recommended solely to build insights into the benefits of ML-assisted RIS in a wireless network setting.