Towards a plug and play population-based structural health monitoring aggregation pipeline design for resource constrained systems

The practical implementation of Population-based Structural Health Monitoring (PBSHM) often involves distributed which face challenges from limited compute resources, power budgets, and variable communication bandwidths, for example, IoT devices with battery-powered wireless sensor network gateways uploading data over metered connections. Building upon the data flow architecture presented in our first paper in this special session, this paper demonstrates how inherent state within data streams' can be leveraged for optimisation. This paper introduces a novel, plug-and-play aggregation pipeline specifically designed to address these limitations. We present an optimised data representation and transmission strategy that minimises computational and bandwidth requirements at the network edge. By leveraging efficient data serialisation techniques, our pipeline achieves a significant reduction in data payload size with negligible information loss, thereby enhancing the scalability and financial viability of PBSHM systems. This work validates an enabling technology for the real-world deployment of large-scale, low-power monitoring ecosystems. It does so by comparing two data formats, JavaScript Object Notation (JSON) and Concise Binary Object Representation (CBOR), using monitoring data from a long-term bridge campaign. The results show a reduction in the volume of transmitted data by up to 12.2 times.