Energy-Efficient Hybrid Active/Passive RIS-Assisted UAV-Enabled IoT Data Collection in Symbiotic Radio Systems

This paper studies a UAV-assisted symbiotic radio (SR) system in which a reconfigurable intelligent surface (RIS) backscatters IoT data to a UAV while simultaneously assisting the primary transmission. To extend coverage without the power and noise penalties of a fully active RIS or the range limitations of a fully passive RIS, we propose a hybrid active/passive RIS that enables element-wise mode selection and per-active-element gain control. We formulate an energy-efficiency maximization problem that accounts for both amplification noise and circuit power under statistical channel state information (CSI), jointly optimizing RIS mode selection, the active-element amplification matrix, RIS phase shifts, the UAV’s 3D location, and transmit beamforming. The resulting mixed-integer, nonconvex fractional program captures tight couplings among geometry, activation, amplifier noise, and circuit power. To solve it, we develop a block coordinate descent (BCD) framework that combines Dinkelbach’s transform for the fractional objective with successive convex approximation (SCA) and a relaxation–rounding strategy for mode selection. Numerical results show consistent energy-efficiency gains over fully passive and fully active baselines (both optimized and random), highlighting the benefits of hybrid selective amplification and UAV placement in SR. We also evaluate a practical discrete-phase hybrid RIS with 4-bit resolution; despite finite-resolution quantization, it closely approaches the continuous-phase design and outperforms the fully passive and fully active baselines.