Abstract

Bats provide important ecosystem services and serve as bioindicators, making their monitoring critical for ecosystem health. However, bat monitoring typically relies on single-sensor acoustic data, lacking environmental context. Moreover, conventional approaches depend on post-hoc analysis, with real-time monitoring remaining rare. A key challenge in multi-sensor integration is the absence of a quantifiable process to merge heterogeneous data streams without compromising fidelity. This research investigates multi-sensor integration for bat monitoring, focusing on fusing acoustic and environmental data using edge computing. The goal is to develop a scalable sensor fusion methodology that maintains data integrity and is translatable to citizen science. This work entails experimental trials on edge computing platforms, producing quantifiable data integrity metrics and replicable integration protocols, and comparing performance under real-world, resource-constrained conditions.