Teresa Vazão, MIAVITA team member,
INESC-ID/Lisbon Tech, Lisboa, Portugal
Deployment of short duration experimental setups with a significant number of sensor nodes dispersed over a wide geographical region is unusual. The reason for the lack of these experiments is because it is too difficult to install/maintain the several nodes or because the deployed equipment is very expensive and risk of damage is high. A promising solution appeared with the emergence of Wireless Sensor Networks (WSN). This technology allowed the ad hoc deployment of low cost systems which can be easily deployed and automatically interconnected, creating networks which are used to transport a broad range of information.
For the MIA-VITA project, the INESC-ID team, located in Lisbon, developed a Wireless Sensor Network solution suited for volcanological experiments.
As shown on Figure 1, the network of the developed solution is composed by 13 nodes spread on the volcano area. The green nodes, named sensor nodes, simply collect and send data. The central node, named the sink node, is a special node as it is responsible for receiving data from all other nodes and relay this information to the Remote Laboratory. The hardware architecture for each node is presented on Figure 2.
It is possible to connect each node to one of two geophone sensors models: uni-axial or tri-axial. Other types of instruments can be used. The measurements reported by these sensors are then acquired by the Data Acquisition System, which is responsible to accurately timestamp the data and to convert the analogue signal to a digital representation.
In terms of processing, each node is equiped with an ARM processor based system, suited for embedded solutions. For communication each node possesses a Wireless 802.11b compatible USB card.
Figure 3 shows the interior of a node. On the right, we can observe part of the Data Acquisition System board, on top of which is fitted the ARM single board computer, running GNU/Linux. To the left we see an internal batery, which provides power for up to 3 hours. On top of it, we can observe a USB thumb drive, for local storage of seismic data.
Using a small team of two or three persons it is possible to deploy the entire network in a short amount of time. As it is possible to see on Figure 4 the developed solution is easy to carry, and installation procedures consist only in securing the geophone and metal case to the ground,
making sure the WiFi and GPS antennea are facing up. No human intervention is necessary for configuring the network, as each node detects the presence of others and the best comunication paths as automatically established.
The WSN can also be deployed for long lived experiments. In this case, external 12V bateries should be used when the experiment is to be run for a few days, or renewable power suplies, such as a solar pannel, can be used when the WSN is to be run continuously.
When data transmission to a Remote Laboratory is not necessary or possible, the sink node can store localy, using an hard drive, data for up to an year. After the equipment is deployed it is possible to check the device status on the terrain either by four status lights present on the metal case, or by connecting via wireless to the node using a normal laptop or compatible tablet. The user in this case is presented with a web page with the node/network status and with the most recent collected samples as shown on Figure 5.