RIS-Assisted 5G and Beyond Communications in Tunnel Environments: Performance Analysis and Optimization

This study investigates the transformative potential of Reconfigurable Intelligent Surface (RIS)-assisted fifth-generation (5G) and beyond, Beyond 5G (B5G), wireless communication systems in the uniquely challenging context of underground mine galleries and other confined environments, such as subway systems, road tunnels, and underground urban complexes. These environments—characterized by severe signal attenuation, complex geometries, and limited line-of-sight (LoS) conditions—demand innovative solutions to ensure reliable connectivity for safety-critical operations and autonomous systems. We evaluate the performance of four distinct RIS phase optimization strategies: Random Phase Assignment, Greedy Algorithm, Semidefinite Relaxation (SDR), and Deep Q-Network (DQN). Using detailed simulations at a 28 GHz carrier frequency, we assess key performance indicators including signal-to-noise ratio (SNR), bit error rate (BER), channel capacity, outage probability, and energy efficiency across varying tunnel distances. The results demonstrate that RIS-enabled intelligent phase control significantly enhances communication robustness and efficiency in subterranean environments, with the DQN-based approach achieving superior SNR gains while maintaining computational tractability. These findings underscore the strategic importance of RIS in advancing resilient, high-performance wireless infrastructures for underground mining and other confined industrial settings.