Publikationsserver
Networked Wireless Sensor Systems for Ecological Monitoring
Due to the climate catastrophe, spatially and temporally comprehensive monitoring of the environment is becoming increasingly important to detect changes quickly. The problem is that human work, due to the limited number of experts, can only be scaled to a limited extent, and humans are not really s...
Gespeichert in:
| Hovedforfatter: | |
|---|---|
| Andre forfattere: | |
| Format: | Dissertation |
| Sprog: | engelsk |
| Udgivet: |
Philipps-Universität Marburg
2022
|
| Fag: | |
| Online adgang: | PDF-Volltext |
| Tags: |
Tilføj Tag
Ingen Tags, Vær først til at tagge denne postø!
|
| Summary: | Due to the climate catastrophe, spatially and temporally comprehensive monitoring of the environment is becoming increasingly important to detect changes quickly. The problem is that human work, due to the limited number of experts, can only be scaled to a limited extent, and humans are not really suitable for performing permanent monitoring. In addition, there are possible cognitive limitations during data collection. If such data is collected almost simultaneously, it can no longer be perceived in its entirety by humans, resulting in data loss. Furthermore, data can be distorted by human assumptions (bias), or observations can be disturbed by the mere presence of humans. Considering these problems, a shift to automated monitoring methods seems inevitable.
This dissertation presents a new approach to create networked wireless sensor systems for ecological monitoring. Especially forested areas often have difficult conditions for humans and material, i.e., the proposed approach must also be able to deal with these real-world problems.
The main field of investigation chosen in this dissertation is the observation of bats. According to the "IUCN Red List of Threatened Species", one-third of them is either considered to be threatened or there is not enough data about them to make a statement. In addition, bats are good indicators of ecosystem health, i.e., monitoring bats can also provide information about the entire ecosystem’s health. Nevertheless, the considerations and practical design ideas presented in this thesis can be applied to other fields of environmental monitoring.
The thesis introduces a novel approach to comprehensively monitor bats, divided into the components of monitoring the ecosystem and monitoring the bats themselves. In these areas, scientific work is presented and concluded with a view of a concrete overall system.
In the area of ecosystem monitoring, approaches on monitoring trees, insects, birds, and human disturbances are presented. In this context, new networked sensor nodes are developed and combined to form a complete system. A novel approach is proposed for equipping existing unchangeable systems and devices with new functions and thus integrating them into existing networked sensor systems. This novel approach is presented using the concrete integration of an existing system into the overall monitoring system.
In the area of bat monitoring, approaches are presented that enable the tracking of bats with different sensors and provide an automatic evaluation. New evaluation methods using machine learning techniques and the development of an operating system for particular sensor nodes are used. As a combination of newly developed hardware and software, a sensor node is presented that enables multi-sensory monitoring of bats triggered by bat activity. As a result of this, video, audio and VHF data can be collected, and ultimately one sensor’s findings can be transferred to different other sensors. |
|---|---|
| Fysisk beskrivelse: | 233 Seiten |
| DOI: | 10.17192/z2023.0108 |