Performance Analysis of IRS-Assisted Networks With Near- and Far-Field Effects

Existing research on performance analysis of intelligent reflecting surfaces (IRS) regularly assumes that the signal transmitted by the IRS is attenuated under far-field conditions. However, the transmission signal shows different characteristics as relative distance varying and the close-range attributes of IRS render existing performance analyses inapplicable. In this paper, a general channel model with a tractable boundary is characterized and the near- and far-field of IRS are distinguished by applying different power scaling laws (e.g. additive and multiplicative path loss) for transmission signal in the channel as well as utilizing Rayleigh distance to demarcate the fields, based on which we analyzed the performance of IRS-assisted networks. Specifically, the impact of IRS element size is modeled into the near- and far-field effects, which directly alters the near-field boundary and influences the receiving signal power of user equipment (UE). In addition, the conditional signal power distribution is characterized as a piecewise function based on the division of Rayleigh distance and the interference Laplace function is rederived accordingly. The numerical results analyze the impact of near- and far-field effects, as well as different IRS parameters, on coverage probability, which demonstrates the significant variation of network performance compared to the far-field assumption when system parameter changes.