Noisy cyclic quantum random walk

We explore static noise in a discrete quantum random walk over a homogeneous cyclic graph, focusing on the spectral and dynamical properties of the system. Using a three-parameter unitary coin, we control the spectral structure of the noiseless step operator on the unit circle in the complex plane. One parameter governs the probability amplitudes and induces two spectral bandgaps proportional to its value. The half-sum of the two phase parameters rotates the spectrum and induces twofold degeneracy under specific conditions. Degenerate spectra yield eigenstates with sinusoidal probability distributions, while non-degenerate spectra produce flat distributions. We introduce static phase noise in the sites and analyze its impact across two distinct propagation regimes. In the walk-on-the-line regime, which precedes a full traversal of the graph, we extract the spreading exponent β from the step-resolved mean squared displacement. We find that low participation ratios correlate with sub-diffusive spread, while high ratios correspond to ballistic or super-diffusive evolution. After the walker completes a cycle, finite-size effects dominate. In this walk-on-the-cycle regime, the spreading exponent no longer characterizes the behavior. We quantify localization using a convergence criterion based on the coefficient of variation of the mean squared displacement. Across both regimes, we identify a sharp crossover near static site noise strength ϕs = π/3. This transition coincides with a drop in the participation ratio, a transition from diffusive to sub-diffusive spread in the walk-on-the-line regime, and a reduction in the saturation level of the mean squared displacement in the walk-on-the-cycle regime. Our results demonstrate that the eigenstate participation ratio provides a computationally efficient spectral diagnostic that anticipates localization across both regimes, offering an alternative to full dynamical simulations.