Data-driven and Model-driven Deep Learning Detection for RIS-aided Spatial Modulation

Jiang Liu (1) , Marco Di Renzo (1)

1 L2S - Laboratoire des signaux et systèmes

Reconfigurable intelligent surface (RIS) is regarded as a key technology for the next generation of wireless communications. Recently, the combination of RIS and spatial modulation (SM) or space shift keying (SSK) has attracted a lot of interest in the wireless communication area by achieving a trade-off between spectral and energy efficiency. In this paper, by generalizing RIS-aided SM/SSK system to a special case of conventional SM system, we investigated deep learning based detection in RIS-aided SM/SSK systems. Based on the idea of deep unfolding, we studied the model-driven deep learning detection for RIS-aided SM systems and compare the performance against the data-driven deep learning detectors.

Domaines: Sciences de l'ingénieur [physics]

Jiang Liu, Marco Di Renzo. Data-driven and Model-driven Deep Learning Detection for RIS-aided Spatial Modulation. 2021 IEEE 4th 5G World Forum (5GWF), Oct 2021, Montreal, Canada. pp.88-92, ⟨10.1109/5GWF52925.2021.00023⟩. ⟨hal-03448167⟩