Abstract
Compared to reflective-only or transmissive-only reconfigurable intelligent surface (RIS), simultaneous transmitting and reflecting RIS (STAR-RIS) has the capability to simultaneously serve users on its both sides. Then, simultaneous indoor-and-outdoor communication (SIOC) has been commonly regarded as one of the most evident and promising applications of STAR-RIS in the next generation wireless networks. However, it introduces intricate interference for wireless networks, i.e., transmission and reflection interference, which would impact the network coverage and transmission capacity drastically. Unfortunately, there is a dearth of research investigating STAR-RIS-enabled SIOC from a network-level perspective. To fill the gap, by leveraging stochastic geometry, we provide the first comprehensive analytical framework for STAR-RIS-enabled SIOC networks. We investigate three commonly used STAR-RIS protocols, i.e., energy splitting (ES), time switching (TS) and mode switching (MS), and consider both unicast communication (UC) and multicast communication (MC) schemes. Exact and approximate expressions of transmission success probability and transmission capacity are derived. Furthermore, we also optimize the key parameters of STAR-RIS protocols to maximize the transmission capacity. Finally, via simulations, the accuracy of theoretical expressions is confirmed, the optimization of the key parameters is investigated, and the advantages of STAR-RIS over conventional RISs are also demonstrated. Moreover, we also conclude the recommended STAR-RIS protocols to UC and MC schemes, respectively.