Reconfigurable intelligent surfaces aided parallel RF and FSO communications using adaptive combing

The channel capacity of free space optical (FSO) communications is sensitive to the atmosphere turbulences, and therefore the application of FSO in the 6G backhaul transmissions faces big challenges, especially in foggy environments. To overcome the channel fading challenge while leveraging the benefits of ultrahigh-bandwidth FSO links, radio frequency (RF) links are still needed, and the hybrid RF and FSO network with parallel links has gained significant attention in recent years. State-of-the-art studies on applying parallel RF and FSO links using hard switching (HS) methods to tackle atmospheric turbulences have shown to be suboptimal if the switching speed is not fast enough to adapt to channel conditions. One effective way to overcome the HS limitations is to utilize adaptive combining (AC) where the FSO link is kept active at all times while the RF link is activated only when the signal-to-noise ratio (SNR) of the FSO link is below a certain threshold value, called AC-Solo. AC-Solo schemes have been studied only for the simple point-to-point direct link cases without relays, limiting its application to 6G network with user mobilities. For the transmitter of a hybrid RF and FSO network to serve receivers at different locations, a relay node using reconfigurable intelligent surfaces (RIS), named RISHS, has also been proposed. However, either AC-Solo or RIS-HS scheme has its drawbacks. In this paper, we propose a RIS-aided parallel RF and FSO network architecture by adopting adaptive combing, called RIS-AC. For the RIS-assisted RF link in RIS-AC, we adopt the Nakagami-m distribution to accurately capture the channel fading effects. For the RIS-assisted FSO link in RIS-AC, we consider various practical factors, including foggy attenuation, pointing errors, and an atmospheric turbulence model using the Gamma-Gamma distribution. At the receiver of RIS-AC, a maximal-ratio-combining (MRC) method is employed to jointly detect the combined FSO and RF signals from the RIS node. To evaluate the performance of our proposed RIS-AC, we first derive analytical expressions for the probability density functions (PDFs) against the end-to-end signal-to-noise ratios (SNRs). Furthermore, we derive closed-form expressions for the outage probabilities (OPs) of RIS-AC based on the PDFs. Simulation results validate our analytical results, and we can see that for an outage probability of 1E-2, RIS-AC has coding gains of 4dB and 2dB compared with AC-Solo and RIS-HS, respectively. The results highlight the superiority of RIS-AC in the hybrid RF-FSO system.