<p>Although base-isolated structures can effectively reduce the seismic response of the superstructure, accurately obtaining their global response after an earthquake is still necessary to support rapid damage assessment. Conventional structural response monitoring usually relies on dense sensor deployment across multiple floors, resulting in high instrumentation and maintenance costs. To address this issue, this study presents a method for accurately reconstructing the global response of base-isolated structures using local monitoring data from the isolation layer together with the ground-motion input. The proposed approach simplifies the base-isolated structure into a multi-degree-of-freedom (MDOF) shear model, incorporating bilinear constitutive laws for both the isolation layer and the superstructure. Next, Particle Swarm Optimization (PSO) is employed to refine the structural parameters based on monitoring data of the isolation layer. A sensitivity analysis is conducted to evaluate the influence of key parameters, providing a foundation for efficient parameter estimation. The effectiveness of the method is validated through numerical simulations, with results showing that the proposed method achieved average Pearson correlation coefficients of 0.95, 0.87, and 0.86 for the displacement, acceleration, and inter-story displacement time histories of each floor, respectively, in the case of a five-story structure. Additionally, robustness assessments reveal that while ground motions and rubber bearing types have a limited impact on accuracy, the number of stories and the choice of optimization function have a greater influence. The proposed method provides a practical framework for reconstructing the post-earthquake response of base-isolated structures under sparse sensing conditions.</p>