Earthquake can cause severe damage to structures, the assessment of structural performance, before, during, and after an extreme event, is critical for ensuring their safe operations and resiliency to potentially catastrophic events. However, the conventional monitoring methods need complicate sensors, acquisition devices and data transmission system, it’s difficult to obtain the real-time response of the structures during earthquake. Moreover, current damage assessment always relies on the numerical simulation, the monitoring data is rare to apply on the damage detection due to the difficult SHM system implementation during earthquake. Furthermore, the displacement was particular difficult to be monitored. In this work, the objective was to extract the damage features such as the modal frequencies variation and residual displacement using smartphone data in a three-story steel frame structure subjected to shaking table earthquake excitations, and study the acceleration integration method in frequency domain to obtain the displacement more convenient and quickly with higher precision. First, a discussion of experimental details, including test structure, test plan and damage cases was introduced. Then the modal frequencies variation and residual displacement in different cases were obtained with the conventional and smartphone monitoring data. Third, the integration techniques for obtaining displacements from acceleration raw data based on one-story measured displacement were investigated to reduce the errors caused by uncertain cut-off frequencies.