Intelligent Reflecting Surfaces for Free Space Optical Communication Systems

• Published in: IEEE transactions on communications Vol. 69; no. 9; pp. 6134 - 6151
• Main Authors: Najafi, Marzieh, Schmauss, Bernhard, Schober, Robert
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.09.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: anomalous reflection; Buildings; Channel models; Communications systems; Electric fields; Equivalence; free space optical systems; Free-space optical communication; Gaussian beams (optics); geometric and misalignment losses; Geometrical optics; Intelligent reflecting surfaces; Laser beams; Lasers; Lenses; line-of-sight link; Mirrors; Misalignment; Optical surface waves; Optical transmitters; Phase shift; Physical properties; pointing error; Reconfigurable intelligent surfaces; Reflection; Statistical analysis; Statistical models; Systems design
• ISSN: 0090-6778; 1558-0857
• DOI: 10.1109/TCOMM.2021.3084637

In this paper, we investigate the use of intelligent reflecting surfaces (IRSs) to relax the line-of-sight requirement of free space optical (FSO) systems. Considering a Gaussian laser beam, we first design a phase-shift distribution across the IRS that enables the reflection of the incident beam in any desired direction, i.e., realizing the generalized law of reflection. Moreover, for the designed phase-shift profile, we show that there exists an equivalent mirror-assisted FSO system that generates a reflected electric field on a mirror that is identical to that on the IRS in the original system. However, the location of the laser source and the properties of the emitted Gaussian laser beam are different in the original and the equivalent systems. This equivalence allows us to study the mirror-assisted system, employing the image method from geometric optics, instead of directly analyzing the original IRS-assisted system. Based on this analysis, we model the geometric and misalignment losses (GML) and characterize the impact of the physical parameters of the IRS, such as its size, position, and orientation, on the end-to-end FSO channel. Moreover, we develop a statistical model for the GML which accounts for the random movements of IRS, transmitter (Tx), and receiver (Rx) due to building sway. Furthermore, we analyze the outage probability of an IRS-assisted FSO link based on the derived channel model. Our simulation results validate the accuracy of the developed channel model and offer various insights for system design. For instance, both our simulations and theoretical analysis reveal that even if the variances of the fluctuations of the Tx, IRS, and Rx positions caused by building sway are identical, their impact on the end-to-end channel is not necessarily the same and depends on the relative positioning of these three nodes.