Springe direkt zu Inhalt

AVS Extrem

AVS Extrem

The German Federal Criminal Police Office (BKA) declares in 2006 a 50-percent increase of the thefts of construction sites. The financial loss for landlords and other owners of construction site equipment was estimated with approximately 90 million Euro. Therefore a flexible, spontaneously installable, economical and interface-open monitoring network is necessary. In this project the Freie Universität of Berlin investigates the possibility to establish distributed event detection in wireless sensor networks by deploying a fence monitoring system as an example.

With the help of a distributed and embedded system, events are to be recognized and evaluated. The challenges of the system are: the wireless communication between the sensor nodes, the event recognition in the sensors data and the design of distributed algorithms for adjustment of the detected pattern. The wireless sensor network (WSN) of AVS-Extrem is a self-configuring and self-healing. It is able to supervise the valuable goods existing on construction sites and to differentiate events from each other.

 

AVS-Extrem  "Autonome Vernetzte Sensorsysteme (mst-AVS)"

Vernetzte Sensorsysteme zur Lokalisierung und Überwachung unter Extrembedingungen
AVS-Extrem-Logo

Partners


BMBF-Logo

Download Flyer <-


Ziel ist die Erkennung und ortsabhängige Bewertung von Ereignissen mit Hilfe eines verteilten und eingebetteten Systems. Neben der Problematik der drahtlosen Kommunikation zwischen den Sensorknoten stellt die Erkennung von Muster in den Sensordaten, die Entwicklung von dezentralen Algorithmen zum Abgleich der erkannten Muster sowie die Einschränkung bei der Hardware wie begrenzte Rechenleistung, beschränkte Energie Versorgung, Speicherplatz und Prozessorleistung eine große Herausforderung dar.

Zusätzlich zur Ereigniserkennung ist das System mit einem kabellosen Ortungssystem ausgestattet, was die 2D-Lokalisierung eines Sensorknotens ermöglicht. Die Kombination von kabelloser Ereigniserkennung und Ortung ist für den realistischen Anwendungsfall der Überwachung sicherheitsrelevanter Areale sehr interessant.

Im Verbundprojekt „AVS-Extrem“ wird konkret ein selbstkonfigurierendes und selbstheilendes Netzwerk entwickelt, welches in der Lage ist, die auf Baustellen vorhandenen wertvollen Güter mit Hilfe sich autonom vernetzender Sensorsysteme zu überwachen und dabei die Ereignisse bezüglich ihrer Alarmierungsrelevanz eigenständig voneinander zu unterscheiden.

Die dafür nötige Erfahrung als auch Technologie bringen die unten genannten Projektpartner in das Verbundprojekt „AVS-Extrem“ ein.

Das System ist somit

  • einmalig bezüglich der Verknüpfung von Abstandsmessungen von Knoten in einem Netzwerk und der Ereignisdifferenzierung der zu überwachenden Objekte,

  • effizient in der Energienutzung und damit langfristig einsetzbar,

  • variable im Einsatz, kann also für viele Anwendungsbereiche eingesetzt werden, etwa auch in Logistikzentren, Lagerhallen und bei Großveranstaltungen,

  • selbstorganisierend benötigt damit nur wenig Zeit und Aufwand bis zum produktiven Einsatz („plug and produce“).

Die Freie Universität Berlin untersucht und in diesem Projekt die Einsetzbarkeit verteilter Ereigniserkennung in drahtlosen Sensornetzen am Beispiel der Bauzaunüberwachung.

AVS-Extrem: A Use Case for Wireless Sensor Networks


The AVS-Extrem project is a use case for Wireless Sensor Networks (WSNs) focused on collaborative, in-network data processing. The goal is to develop a distributed event detection algorithm that can reliably report security relevant incidents (e.g. a person climbing over a fence) to a base station. The vision is that through cooperation of many sensor nodes the accuracy of event detection can be greatly improved, while at the same time saving energy by reducing multi-hop communication with the base station.


Distributed Event Detection

Layered Event Detection Architecture



We have taken a layered approach to tackle the problem of in-network event detection, thereby isolating the subproblems of per-node event detection and coordination between neighboring nodes. A diagram of our architecture is shown below.


Layered Event Detection Architecture

The different layers of our architecture implement a distributed, multi-step event detection algorithm. In the lowest layer, raw sensor readings are isolated from background noise and aggregated into a set of characteristic properties. The next layer checks whether known patterns appear in these aggregated values and identifies them as event candidates. In the next layer, the sensor nodes collaboratively decide whether a noteworthy event has in fact occurred within an n-hop neighborhood by exchanging information about recently observed event candidates. Finally, the uppermost layer reports confirmed events to the base station of the deployment.


Deployment Videos



In September 2006, we evaluated our event detection algorithm with a small deployment of sensor nodes attached to a construction fence in the patio of our institute. The video below illustrates the types of events we considered in our experiments.

   

In October 2008, we once again evaluated our reworked event detection architecture on the example of a wireless alarm system. 100 sensor nodes equipped with accelerometers were attached to the fence surrounding a real-world construction site. The task of the WSN was to detect security relevant incidents by recognizing four previously trained patterns in the lateral oscillation of the fence elements. As above, this video illustrates the types of events we considered in our experiment.



In
August 2011, we integrated our distributed event detection in a wireless motion-based training device. We showed by the example of Martial Art Stick Fighting that the event detection is able to support people during performing movements. This system may also be useful in rehabilitation scenarios or other wireless body area networks (WBAN). The video illustrates the functionality of the device by performing six simple stick fight techniques. 

 

Relevant Publications


 

Contact: Norman Dziengel