OASIS System Configuration
Figure 1 illustrates the end-to-end configuration of OASIS system. The ground network delivered realtime volcanic signals to the sink nodes through multihop relays. The gateway (MOXA device server DE- 304) relayed the data stream to WSUV through a microwave link of 50 miles. In the lab, a customized TinyOS tool SerialForwarder forwards the data between the sensor network and the Internet. Multiple control clients may connect to it, access the sensor data stream, and control the network in real time. V-alarm is a volcano activity alarm system, which can automatically identify earthquake events from the raw data stream. Once an event is triggered, V-alarm can send event alerts via email or text messages to the corresponding scientists in charge. The Command & Control center is for situation awareness and integration of in-situ sensor network and space observations from EO-1 satellite. It incorporates existing real-time volcano monitoring and data-processing tools used by the USGS and makes realtime autonomous operational decisions to control the sensor network according to local and remotely sensed environmental changes.
To meet the needs of monitoring a harsh environment and support air-drop deployment, we designed a 3-leg spider station (see Figure 2). To enable sensor nodes to connect with each other, we have mounted a 6 dBi omnidirectional antenna on a PVC steel pipe to get reasonable line-of-sight.
Inside the spider, the core hardware components are encapsulated in a small weather-proofing sensor box, with a dimension of 30×20 cm² (see Figure 3). The sensor box contains a sensor mote (iMote2), a sensor board (MDA320CA), a customized 802.15.4 amplifier, a GPS receiver (LEA-4T) and other components.
Each sensor node contains a seismometer to detect earthquakes, a GPS receiver to pinpoint the exact ocation and measure subtle ground deformation, an infrasonic sensor to detect volcanic explosions, and a lightning sensor to detect eruption clouds.
Figure 4 illustrates the architecture of the software architecture inside sensor nodes.
The real-time data stream from seismic, infrasonic, lightning, and GPS sensors, as well as RSAM, battery voltage and RSSI/LQI data, are imported into a MYSQL database with GMT timestamps of millisecond resolution. The Figure 5 illustrates the network and data management tools. ExportOasis categorizes the data stream into different types such as the seismic, infrasonic, lightning, battery voltage, LQI. Wave data (seismic and infrasonic) is reformatted into standard form for storage by ImportEW before logged into the MySQL database. Our database is well integrated with USGS’s science analysis software (e.g. VALVE) to manage and visualize the volcano monitoring data. VALVE (Volcano Analysis and Visualization Environment) is a on-demand client/server system for visualization of serving, graphing, and mapping nearly every type of historical data collected by the sensor network. It greatly facilitates our analysis of the data quality and network status. Our data is shared with the community through the VALVE web client. It allows users to view or download the volcanic data of a specified period from any location on the Internet.