Towards Building IoT Tools and Services for Earth Science Community

Abstract:

The Internet of Things (IoT) is a system of interrelated computing devices, mechanical and digital machines/sensors with unique identifiers (UIDs) having the ability to transfer data over a network without human intervention. Although great potential exists for earth system science community to take advantage of the IoT technologies, case studies are rare. We recommend developing an interdisciplinary software platform using IoT technologies for earth science data collection, storage, integration, and processing. We implemented a specific instance of this platform that monitors various components and processes of an urban forest ecosystem located in Hunan province, China. Various sensors have been installed in the forest to detect the behaviors of animals and plants as well as environmental conditions such as soil and air temperatures at different locations in the forest canopy and soil. Users can visually see these data and information on the software platform. Combined with AR technology, users can also directly scan the IDed objects through mobile devices to find detailed information of the objects on demand. Various classifier neural networks, independent of the specific network structure and training algorithm, have been deployed on the platform to facilitate the analysis of the IoT. Embedded devices adopted the physical unclonable function (PUF) technology for data security, and the neural network encryption algorithm of the IoT used the Advanced Encryption Standard (AES). This implemented instance is scalable and can be easily extended. In ideal condition, the large quantities of science data collected from disparate hardware equipment can be processed in the visual software platform, which will then apply simplest and most understandable pages to help interdisciplinary researchers.