Sensor

1
Sensor Computing Infrastructure for
Environmental RisksSCIER FP6-2005-IST-5

• Stathes Hadjiefthymiades (NKUA)
• 1st Student Workshop on Wireless Sensor Networks
• Brussels, Oct. 2008

2
Main Objectives

• Sensor network infrastructures for the detection
  and monitoring of disastrous natural hazards.
• Advanced sensor fusion and management schemes.
• Risk evolution models simulated on GRID.
• Multi-risk platform.
• Public-private sector cooperation.

3
SCIER architecture
4
SCIER Sensing Subsystem

• Sensor Infrastructure
• In-field sensor nodes (humidity, temp, wind
  speed/direction)
• Out-of-field vision sensors (vision sensor)
• Sensor Data Fusion

5
SCIER Computing Subsystem

• Computation and Storage
• Environmental models
• Flash Floods (FL), forest fires (FF)
• GIS Infrastructure
• Storage, analysis and visualization of monitored
  data, spatial calibration and event localization
• Predictive Modeling
• Front-End Subsystem

6
Local Alerting Control Unit (LACU)
Alerting Infrastructure
Sensor Infrastructure
Computing Subsystem
Sensing system proxy
LACU Software modules
XML
OSGI
JDBC
DataBase
Data flow
Control flow
7
Computing Subsystem Architecture
8
LACU Fusion Component (FF)

• Receives sensor data and executes fusion
  algorithms.
• Generates fused data with degree of reliability.
• Fused data fed to the Computing Subsystem.

9
2nd Level Fusion Process (FF) in CS

• Camera data and Fused sensor data from LACUs are
  processed .
• Algorithms
• Voting algorithm
• Dempster Shafer Theory of Evidence
• Triggers simulations according to the final
  probability of fire, flood, etc.

10
FF simulation modeling

• Simulation of several possible futures through
  the GRID infrastructure.
• GRID used to simulate many possible future
  situations (1-100) under different propagation
  conditions
• results analyzed to identify the size and shape
  of the resulting burned area, and provide
  probabilities for each of the simulated futures.

11
FL Modeling

• Conditions stored in metadata catalog
• Engine for parsing and evaluating conditions
  based on incoming data.
• Interface with Simulation subsystem triggering
  model execution based on fusion result

conditions
Metadata Catalog
Sensor input data
Condition evaluation engine
Fusion Decision
12
SCIER GRID and FL with web-services
Sensors

• Collect data (locationtimevalue)
• precipitation
• temperature
• humidity
• wind

Fusion
SCIER central point
User interface
GRID
Forwards data to storage Issues simulation
jobs Runs web server with UI
Executes fire modelling jobs
HTTP
ArcGIS
File share, SQL
SQL
Web services
Services

• Storage for
• fire models executables
• model input data
• model structural data
• model output data
• Pre-prepared WS CS scenarios

Simulation PC(s)
Executes 1D flood modelling jobs Incorporates
pre-calculated flood maps lookup
13
System Validation Evaluation

• Testing includes both fires and flooding
• Gestosa, Portugal (experimental and controlled
  burns)
• Stamata, Attica, Greece (fires, system deployed)
• Aubagne, Bouches du Rhone, S. France (fires and
  floods, deployment underway)
• Brno, Czech Republic (floods, system deployed)

14
System Validation Evaluation

• Gestosa, Portugal (experimental and controlled
  burns)

15
System Validation Evaluation

• Stamata, Attica, Greece (fires, system deployed)

16
System Validation Evaluation

• Aubagne, Bouches du Rhone, S. France (fires and
  floods, deployment underway)

17
Thank you!
Project website http//www.scier.eu