← Back to Sessions

Automated Operational Modal Analysis of High-rise Buildings Using Crowdsourced Smartphone Data

Description: 

Quantifying the dynamic properties of urban infrastructure is fundamental to seismic risk assessment, yet the vast majority of the global building stock remains uninstrumented. We present a scalable computational framework for estimating fundamental building natural frequencies using crowdsourced ambient vibration data from the MyShake smartphone network. To address the inherent challenges of consumer-grade sensors and arbitrary device placements, we developed a rigorous signal processing pipeline. This methodology integrates geospatial association to link devices to building footprints, followed by polarization analysis to autonomously recover principal structural axes without reliance on magnetometers. By aggregating short-duration recordings, we extract stable fundamental modes through automated spectral clustering. The robustness of this framework was evaluated against scientific-grade instrumentation during a full-scale shake table test of a 10-story cold-formed steel framed building. The results demonstrate that smartphones can reliably capture key dynamic parameters, effectively transforming personal devices into a dense, ubiquitous sensor network for the continuous characterization of city-scale structural performance.

Session: Earthquake Ground Motions and Structural Response: Emerging Tools and Applications - II

Type: Oral

Date: 4/16/2026

Presentation Time: 10:30 AM (local time)

Presenting Author: Utpal Kumar

Student Presenter: No

Invited Presentation: 

Poster Number:

Authors