Designing STAR-RIS-Assisted Wireless Systems With Coupled and Discrete Phase Shifts: A Computationally Efficient Algorithm

The concept of simultaneously transmitting and reflecting reconfigurable intelligent surfaces (STAR-RISs) has been proposed to create an intelligent radio environment in full space. However, previous studies have assumed independently adjustable phase shifts, which can be complex and limit scalability. Recent research has revealed that low-cost passive lossless RISs have coupled phase-shift coefficients for transmission and reflection, presenting new design challenges. To tackle these challenges, we propose a novel algorithm for STAR-RIS-assisted wireless systems with coupled and discrete phase shifts. Our algorithm uses zero-forcing beamforming for transmit beamforming and employs a probability learning technique based on the cross-entropy optimization framework to determine phase shifts and corresponding transmission and reflection amplitudes. Simulation results demonstrate that the proposed algorithm achieves higher sum-rates and lower computational running times than existing methods.