Thawing permafrost in alpine regions poses an increasing threat to human life and infrastructure. We maintain a wireless sensing system which enables long-term and multi-modal monitoring of mountain slopes in harsh, high-alpine regions. To acquire as much information as possible under the constraints given by this environment we develop and optimize learning-based algorithms for the edge and for the cloud.
Sorting relevant from irrelevant information on the edge reduces the transmission cost and increases reactivity while cloud computing permits a qualitative, long-term analysis.
By taking the whole chain into account from low-power embedded device via data transmission and processing to data visualization and analysis we present an optimized tool suite for monitoring the effects of climate change in steep permafrost rock.

Download the slides for this talk.Download ( PDF, 13826.91 MB)