Impact of Wind-Induced Phase-Shift Errors on RIS-Assisted Cell-Free Massive MIMO Networks: Spectral and Energy-Efficiency Analysis

Reconfigurable intelligent surfaces (RISs) can improve spectral efficiency (SE) and energy efficiency (EE) in cell-free massive multiple-input multiple-output (CF-mMIMO) networks by coherently controlling reflected propagation paths. Outdoor panels, however, can experience wind-driven displacement and vibration that perturb their programmed reflection phases. This study develops a physics-informed, explicitly phenomenological framework that maps wind speed to the standard deviation of RIS phase error and propagates the resulting uncertainty through a downlink CF-mMIMO simulator. The mechanical chain is made explicit through aerodynamic dynamic pressure, panel displacement, electromagnetic path-length change, and element-wise phase error. Four operating scenarios are evaluated: no RIS, ideal RIS, a wind-disturbed RIS with stale channel state information (CSI), and a robust RIS with reduced residual phase error and refreshed CSI. A two-factor ablation separates the effects of residual phase error and CSI freshness. Under the nominal conditional model at 28 GHz and 21 m/s, the disturbed scenario produces 20.67% lower SE and 20.67% lower EE than the ideal-RIS reference. The ablation confirms that the transition from the disturbed to the robust scenario cannot be attributed to mechanical stabilization alone because CSI refresh and its interaction with the residual error also affect the result. Additional analyses examine coefficient uncertainty, placement, direct-link blockage, carrier frequency, per-user fairness, RIS size, spatial correlation, transmit-power compensation, and control-overhead sensitivity. The nominal robust-case EE excludes implementation-specific stabilization and sensing hardware; an explicit overhead sensitivity is therefore reported. The results should be interpreted as conditional system-level evidence, not as an experimentally validated wind-to-phase calibration law. A wind-tunnel and over-the-air validation protocol is specified for future calibration.