HUMBOLDT-UNIVERSITÄT ZU BERLIN
COMPUTER SCIENCE DEPARTMENT
Systems Architecture Group

School destroyed by an earthquake in El Salvador (www.usaid.gov/sv/)

The cost of natural disasters - lives lost, homes destroyed, economies disrupted - has exploded in the last century. While there has always been natural phenomena like earthquakes, flooding or droughts, nowadays more people are affected because Earth’s population has increased. Early warning systems detect such upcoming phenomena and help to save lives and property.

Geo science applications like earthquake early warning, but also prediction of weather or measurement of other environmental factors are based on the availability of sensor data. The more data available from different locations and different points in times, the more accurate the results. In the past the collection and communication of the data (the geo data infrastructure) has been a complex and expensive task. Geo science applications therefore either tended to be extremely expensive, or restricted themselves to relatively few sensors, collecting the data only sporadically.

Applications like earthquake warning systems however require data from large amount of sensors being processed in real time. If a local earthquake is detected by only few, widely distributed sensors the warning might not be reliable. If communication and processing of the data takes too long, the warning will be late. Similar constraints apply to other warning and monitoring systems as well.

Recently, inexpensive high throughput wireless communication hardware became available. Using these devices, wireless networks can cover large areas, allowing applications to access data from large amounts of sensors in real time. Data can be furthermore aggregated and processed within the network, thereby reducing the communication overhead and adding to the robustness of the system.

The System Architecture Group researches wireless communication protocols and technologies for use in distributed geo science systems. We use inexpensive 802.11 WLAN devices to built robust, self-organizing networks that are easy to deploy. Our current focus is on systems for early warning and geo monitoring.

The graduate school METRIK is a joint project of five institutes, sponsored by the DFG. Focus of the research is to investigate model-based development of self-organizing information systems in catastrophe management. Twelve researchers from computer science and geography are conducting basic research in applying workflow management technology to disastrous events, such as earth quakes, based on self-organizing network protocols and information service concepts.

Partners: GFZ Potsdam, HPI, Fraunhofer First, Zuse-Institute Berlin

---

SAFER is an international project to develop effective earthquake early warning systems in Europe and, particularly, in Europe's densely populated cities. The project is funded by the sixth framework programme of the European Union, and involves 23 institutions from 14 different countries.

Partners: GFZ Potsdam and others

---

Objective of the EDIM project is to develop an earthquake disaster information system for the Marmara region in Turkey. The already existing Istanbul Early Warning System is extended in terms of area, methodology and data transmission. Novel technologies in seismology, structural engineering, and information technology are developed and tested. The project is funded by the German Federal Ministry of Education and Research.

Partners: GFZ Potsdam, Universität Karlsruhe, KOERI, CEDIM

---

The BERLEX project measures temperatures, humidity and wind speed at various outdoor and indoor places in Berlin. We feed the raw sensor data into a geo-model which can identify areas with severe conditions and high potential health risks for certain population groups, e.g. elderly people, small children, people with high blood pressure.

Partners: HU - Geografisches Institut

---

The Dhaka project develops strategies to improve the quality of life in mega cities like Dhaka, Bangladesh. One of the project's goals is to monitor different environmental factors like air pollution and heat in the cities.

Partners: HU - Geografisches Institut, Universität Bielefeld, Universität Freiburg, Leibnitz-Institut for Freshwater Ecology and Inland Fisheries, Rajshahi University