Geometric Confinement of Decoherence in Connectivity-Engineered Networks

The core challenge in quantum computing lies in decoherence and thermalization caused by environmental noise. Within the framework of connectivity engineering, this paper proposes a protection scheme based on pure geometric decoupling mechanisms. By separately calculating the dynamical evolution of both closed and open systems, the numerical results show that: in a closed system, the local network (k=1) thermalizes rapidly, while the decoupled network (k=3) exhibits long-time quantum oscillations. Under Lindblad dissipation, the Loschmidt echo of the k=1 network decays rapidly; while under the parameter conditions of L=10 and γ=0.5, the fidelity of the k=3 network still revives to 0.41 at t=12.5. This mechanism originates from geometric truncation, which fragments the long chain into independent sub-chains, forming "geometric confinement" and "Quantum Fast-Shutter Protection". After further introducing static disorder impurities, the revival peak remains stable, indicating that the mechanism possesses strong robustness against both dynamical dissipation and static disorder. This result establishes a connectivity criterion for suppressing decoherence and provides a solid discrete physical benchmark for designing noise-resistant NISQ quantum memory hardware.