Sensor Service Grid as Rea-Time Monitoring Infrastructure and its Applications in

13 July 2009 University of Tokyo

HONDA Kiyoshi, Asian Institute of Technology / Mie University,

Aadit Shrestha, Apichon Witayangkurn, Rassarin Chinnachodteeranun, Supasak Kulawonganunchai Asian Institute of Technology, Thailand

Hiroshi SHIMAMURA elab experience Inc., Japan

Contents

Background Sensor Asia Sensor Service GRID ( SSG ) Applications

1 Ubiquitous Geo-Informatics for Modeling & Simulation

RS: Regional-Global Sensor Network

Web GIS for Data Sharing

Super ComputerCompa ris on of Sate llite LAI an d Simula ted L AI 5 Model LAI_s at 4 LAI_s im 3 Calibration I A

L 2

1

0 0 30 60 90 120 150 180 210 240 270 300 330 360 Model DOY Scenario Simulation Real Time

Field Server Sensor Network

Web Server Wi-Fi or Cellar-phone A/D Sensors Camera Prof. Masayuki HIRAFUJI, LED Lighting NARO, Japan Solar-cell

2 Field Server Network Real time Data Collection

Sophisticated Interface

3 Field Sensor Network’s High Potential Low Cost Sensor Field Sensor Network Platform Field Server ( NARC, elab experience…) Meshnet Mobile Internet Mobile Phone (GSM, 3G) , Internet Satellite(IP-STAR) WiFi ( Hotspots, Green-WiFi in Bangkok ) New generation of mobile internet ( WiMax ) Applications Environmental Monitoring Agriculture Logistics Security Problems Setup/Configuration Cost Metadata Management Remote Management Connection to Real Application Operational System

Sensor Asia Initiative

4 Sensor Asia Promote High Density Field Sensor Observation

Real Application R&D Automatic Setup Manual Setup Sensor Plug&Play Know ledge on Netw ork, Database….

Obtain Real Time Field Data Globaly High Potential for various Application

FS Specialist Farmer, Application Engineer, Computer / Network Engineer Kids interested in Environment

Sensor Service GRID (SSG) Sensor Manufacturer Sensor Observation Service

Sensor Service GRID

5 SOS: Sensor Observation Service OGC: Open Geo-Spatial Consortium Standards on SWE SOS ( Sensor Observation Service ) SPS ( Sensor Planning Service ) WNS ( Web Notification Service ) Obtain Sensor Information Standard Query and Answer O&M Documents ( xml ) Sensor Name, Location, Spec, Data….. Provide standard interface to sensor data Figure from http://52north.org Metadata( Information on Sensor ) , Sensor Data can be obtained by standard query to the system Improvement for practical deployment Field oriented Fill-up Gap from physical sensor setup to SOS system Security Public Service

SOS Station Communicate via Standard Protocol Returns Standard O&M Document

+ =

Linux Box

Fieldserver SOS Implementation SOS Station

6 Sensor Service GRID (SSG) Simple Overview SOS Stations ( Field Sensor Node )

Visualization, Local Client SOS

Slow, Unstable Network Firewall, NAT

Global Client Visualization, SOS, On/Off-Line Node- Management Naming, Grouping, Monitoring, Archiving, Controlling, Publishing Control, Visualization, GML/KML, SOS User Management, Security Applicati ons Agriculture M anagement Application Development Disaster Information Dashboard Environment SSG Servers Energy Monitoring ( JAVA/API, Mashup, Flex, AIR.. )

SSG GIS & visualization interface

SensorAsia provides a user-friendly GIS visualization interface Shows locations of Fieldservers on web-maps Shows the data from various kinds of sensors and weather- stations in easy-to-understand graphs and dials.

7 Global Access Global access Anytime Anywhere Reliable Stored Data on remote and central server Reduce maintenance cost (engineer can add/remove/edit configuration from anywhere) No Server system by users

SOS Station Management Status & Grouping

SOS Station Status Monitored: blue Not Monitor: gray

SOS Station Group Owner can manage group by themselves Grouping by Company Location Type of system

8 Open System for Sensor Connection Open to Sensor Manufacturer Small Feeder Program Obtains Data from Physical Devices Data feeder template for Sensor manufacturer for quick connection Sensor Manufacturer can promote their sensor ; No need to take care of data archiving/publishing process Supports Multi Devices Field Server III, IV Davis, Lacrosse Decagon SenseAir I/F http, serial, DB

Sensor Plug and Play Easy to add new sensor to SOS station

Select a Sensor Model

Edit Sensor Spec

Add Sensor

Select Sensor Model and Edit Configure Sensor Setting even Remotely The system automatically starts archiving, changes interface. No Programming Required -> Reduce Cost

9 Open System for Application

Open to Solution Providers 3rd Party Application can be easily constructed on SSG Int’l Standard I/F – SOS Java API for SOS query and IDE Chronograph Layer for Data Filtering Applications send queries to SSG and obtain Answer ( Sensor Spec, Sensor Data ) Construct Purpose Oriented Application : Multi-Site Viewer, Early Warning, Simulation and etc.

IDE and 3rd Party Application

Selecting Sensor and Adobe AIR Application connecting to GUI

10 Spinach Field Monitoring using FS For Food Safety Sensor Asia To promote high density field sensor network for various issues Agriculture, Food Safety, Disaster, Environment… ALFAE ( Area-wide e-Laboratory for Food, Agriculture & Environment ) Pesticide Residue Shift spinach supply to Thailand SWFIT Co., Ltd. – University COOP Follow Euro GAP ( Good Agricultural Practice ) Deliver real time image to Consumer ( Univ. COOP’ Restaurants ) Promote products and contribute income generation Foster confidence among farmers/students on ICT Bridge between Japan and Thailand DIAS Data Integration and Analysis System Project, www.diasjp.org

Spinach Field Location Spinach Field Chaing Dao, Chaing Mai Chaing Mai 3 hours from Chaing Mai Elevation 1200m Thailand

Bangkok

11 Setting

12

Davis d l Weather Yagi e Station i Anetenna F h

c FS Engine

a Camera

n CO2 i Sensor p S t Omni Antenna WDS AP Davis Console a

for School and r Research HUB e Center Linux Box v r e Soil Moisture S

Temp. Conductivity d Leaf Wetness l Sensor Data Logger e i Soil Heat Flux Sensor F

SSG Demo

13 WVIEW Webpage http://203.159.10.20/weather/ChiangMai/

WVIEW Webpage

14 ECH2O-TE 32cm

WVIEEWC WHe2bOpa3g2cem Soil Moisture (mV)

15 Em50 Battery Monitor

HeatFlux & Co2 (via FieldServer A/D)

16 Weather Station Data (Davis)

Spinach Field Images

Harv esting in mid of 2008/01

17 Embedding Soil Moisture Data to Your Webpage

Glacier Monitoring in Himalaya

Global Warming Melting Glacier -> Glacier Lakes GLOF ( Glacier Lake Outburst Flood ) High Potential to Create Debris Flow to Rural Community and Tourist Monitor Glacier Lakes using FS for Early Warning Imja Glacier Lake

18 Fieldserver near Imja lake (5000m)

Fieldserver @Island Peak Ridge (5200m)

WiFi Relay Station at Chukkung-Ri (5100m), relaying 23 Km to Namche

Glacier Lake Monitoring, Imja Lake (5000m), Everest Region,

Some Definitions

Glacial Lake A glacial lake is a lake with origins in a melted glacier. (http://en.wikipedia.org/wiki/Glacial_lake)

©Samjwal Bajracharya ICIMOD

19 Imja Lake (Imja Tso), Ev erest Region

Some Definitions

Glacial Lake Outburst Flood (GLOF) GLOF is technically a sudden and often catastrophic flood that occurs when a lake contained by a glacier or a terminal moraine dam fails This can happen due to: Erosion, water pressure, avalanche, earthquake, or large portion of glacier falling and massively displacing water from the lake.

20 In Perspective …

GLOF can cause massive damage to human settlements in lower areas Past records of huge loss of life and property around the world Need to identify critical glacial lakes and monitor them Develop early warning systems Imja Lake is classified as highly critical

Study Area

Imja Lake Non-existent in 1960 Started as a small pool of water Now covers more than 1 Km sq. Highest rate of retreat for a glacial lake in the , 74m per year (between 2001 and 2006) (ICIMOD-UNEP Report, June 2007)

21 Travel Itinerary

17-28 Sept, 2007

Travel Itinerary

22 Implementation Outline

Link from Namche Bazar seems to be the best option A plan has been received from NREN (Nepal Research and Education Network) A wireless link with 2 hops – Wireless Relay stations at 2 places Lake<->Chukkungri<->Kongde<->Namche

Implementation Outline

© Mahabir Pun, NREN

23 Implementation Outline

© Mahabir Pun, NREN

Implementation Outline

Nepal Wireless Project, Annapurna Region http://www.nepalwireless.net

24 Draught Monitoring in Thailand Big damage to agriculture Damage Assessment and Prediction Modeling and Simulation Dynamic Water Balance Flux Observation Soil Moisture Access to Data through SOS Thai Research Fund

Soil-Water-Atmosphere-Plant Model (SWAP)

Adopted from Van Dam et al. (1997) Drawn by Teerayut Horanont (AIT)

25 SWAP Model Parameter Determination - Data Assimilation using RS and GA - SWAP Input Parameters sowing date, soil property, Water management, and etc. RS Observation

SWAP Crop Grow th Model

LAI, LAI, Ev apotranspiration Evapotranspiratio n I

I 4.00 4.00 A A L L Fitting n n

o 3.00 3.00 o i t i t a r a i r p 2.00 Assimilation by i 2.00 s p n s a n r t

finding Optimized a o 1.00 r 1.00 t p o a

v parameters p v E

0.00 a 0.00 0 45 90 135 180 225 270 315 360 By GA E 0 45 90 135 180 225 270 315 360 Day Of Year Day Of Year RS Model

Impact of Draught

Accumulative AnAnunanl Ruaailn fRalla iinn Ubon Ratchathani 20 00.0 19 90.8 1874.2 1790 .5 18 00.0 ) . 158 1.7 1622 .3 1581 .4

m The lowest rainfall 16 00.0 14 83.2 1484 .6 m (

14 00.0 t 1273 .7 n

u 12 00.0 o

m 10 00.0 A

l appeared in 2003 8 00.0

a Lowest f

n 6 00.0 i a

R 4 00.0 2 00.0 but the most 0.0 20 01 2002 2003 200 4 20 05 20 06 2007 2008 Avg.19 71- 2003 Year 200 0 serious impact on

Averaged RYicee Yiliedld iSn tUabotnis Rtaitchsathani Province rice yield was 430 421 413 420 406 410 400 400 ) i a 400 R / 387 found in 2004 g 390 Lowe ( K 372 d

l 380 e i

Y 370 st 360 350 October Rainfall 340 2001 2002 2003 2004 2005 2006 2007 20Y0ea 4r 2003 43.6mm Yield SimulatSioimnu laUtend rdicee ry ieDld iufnfdeer e1n mto nDth rdyry Ssppelel ll Scenarios 2004 3.3mm

400 344.80 349.12 363.04 352.96 353.28

. 350

Dry spell in

300 Yield Simulation with ) i a 250 R / Dry Spell in October

g October has 200 K

( 151.68

d 150 l e i 100 Y serious impact 50 0 July August September Oct ober November Actual yield

26 DIAS

地球観測データ統合・解析システム Data Integration and Analysis System 国家基幹技術 衛星観測データー＞社会に有用な情 報へ Modeling / Calibration / Validation Khon Kaen, Thailand; Soil Moisture, Agriculture Monitoring Installed in Dec 2006 Constructing Data Path Khon Kaen -> AIT (SSG ) -> UT

Soil Moisture EC-5 Khon Kaen

27 Asian Joint Research Project for Early Warning of Landslides Joint Project by , Japan , Korea, Philippine, Thailand Banjarnegara Landslide ( slow movement type ), Jogjakarta, Indonesia To propose early waning system of Landslide for Asian Implementation Installing Field Server for Early Warning Dec 2007 Water Pressure, Extensiometer, Rain gauge and Camera Early Waning Local / Remote Monitoring

28 Remote Monitoring System Configuration

Outdoor Unit AIT Server (FS+Sensors) 2 Extensiometers Rain Gauge WiFi, GPRS Water Pressure Cable Camera

ter2 Indoor Unit Me sio Local Serv er ten Ex GPRS Modem Speaker UPS 2nd Pole ter1 Me sio 3rd Pole ten Ex

Local Monitoring/Warning

Pulley for Extensiometer Invar Line Rain gauge, IP Camera and Pulleys on the 2nd Pole

Invar Line to Pole3 IP Camera

Invar Line to Pole1 Rain Gauge

Tension Weight to Invar Line Invar Line to extensiometer

Extensiometer

29 Out WDS AP HUB

Door DC Power Supply Unit Rain Gauge Counter Water Pressure Extensiometer Strain Gauge Amp Amp for rotary potentiometer ( 2 potentiometer with 180deg phase diff)

New Fieldserver Engine NARC 24 bits Thanks to Prof. Hirafuji

Water Pressure Sensor (Strain Gauge) Depth at 2.51m

Indoor Unit Speaker Local Server UPS with External LINUX Box Battery With IM GPRS Modem

30 Warning I ( First Warning, Yellow ) R24>100mm && R24 > 250m3m -LR7e2vel of Warning Warning II ( Prepare for ev acbuatsioend, Oorann gAe )ntecedent Warning ( R24 > 150mm && R24> 350mm-R72 ) || Antecedent Rain ( Warning I && ( e1 > 2mm/hRr oar ien2 f>a 2lml ma/nhrd ) ) Screen 10min, 1 hr, 24 hrs, Warning III ( Ev acuation, RedD) isplacement 72hrs If ( Warning I || Warning II ) && Warning ( e1 > 5mm/hr || e2 >5mm/hr ) Meter R24, R72: antecedent rainfall in 24 , 72 hours Rain fall in 24hrs e1, e2: displacement rate Important to provide understand able information to people 60mm in 20 hrs !

Displacement (mm) in 24 hours Displacement rate (mm/hr) in 4 hrs

Water Level(m)

Exhibition to HRH Princess

12 Nov 2008

31 Future Plan

Public Service Beta 1 Testing since 24 April 2009 is going on Commercial Service, September 2009 Sensor ID For Sensor P&P Develop various applications and testing Thailand Disaster Management System Building Energy Monitoring System Nepal Aviation Route Monitoring Food Safety ( Sensor Data -> Food Safety System, GAP ) And etc. Gateway <-> Metbroker GIS Mashup Service

Conclusion Sensor Service GRID as an infrastructure Standard Protocol, Global Access, Sensor Plug&Play Automate transfer, meta data management, publishing Low Cost Installation Sensor middleware for various businesses Various Application Public Service Commercial Service: September 2009

32 Development Team

Mr. Apichon Witayangkurn Ms. Rassarin Ch.

Mr. Aadit Shrestha Dr. Honda Kiyoshi Mr. Supasak Kulawonganunchai

Thank you

[email protected]

CO2 Sensor Demo

patent pending

33