A Framework for Mobile Crowdsensing in Bridge Health Monitoring

Mobile crowdsensing (MCS) represents a promising solution for the widespread monitoring of bridge infrastructure, which allows continuous data collection across large networks. Despite its considerable potential, the systematic application of MCS to bridge structural health monitoring remains largely unexplored, with no established framework to guide implementation or address the associated technical challenges. This paper places particular emphasis on data‑quality screening, a critical yet underexplored requirement for ensuring the integrity of crowd‑submitted measurements. Screening quantity is proposed to assess submission quality and enforce quality‑aware data selection. The framework is validated using acceleration recordings that include both admissible and inadmissible submissions, demonstrating reliable discrimination between high‑quality and low‑quality data. The proposed quantity is computationally lightweight and designed for on‑device execution, which allows only accepted recordings to be uploaded to the centre and substantially reduces bandwidth and storage demands. It also supports a quality‑aware incentive mechanism that encourages sustained participant engagement and progressive improvement of data quality over time. Additionally, a quality‑aware data‑fusion procedure is introduced, and the complete methodology is validated using benchmark bridge data.