End-to-End Fading Channel Modeling for RIS-Empowered Smart Wireless Environments

The combination of reconfigurable intelligent surfaces (RISs) and extremely massive antenna arrays is lately gaining increased attention as a new wireless networking paradigm, offering programmable radio wave propagation and holographic communications. However, the channel models used to date for the design and analysis of RIS-empowered multiple-input and multiple-output (MIMO) communication systems include fading in a manner that is unfaithful to the underlying wave physics. In this paper, we introduce a physics-based end-to-end model of RIS-parametrized MIMO channels of arbitrary dimensions with adjustable fading (named PhysFad), which is based on a first-principles coupled-dipole formalism. PhysFad naturally incorporates the notions of space, causality, frequency selectivity, and the intertwinement of phase and amplitude response of each RIS and antenna element, as well as any arising mutual coupling effects including long-range mesoscopic correlations. The presented channel model, whose code is shared openly, is used to implement Ricean fading conditions with various K factors in a physics-compliant manner.