Details of the North Temperate Lakes Long Term Ecological
Research Site's Wireless System
Contact: Tim Kratz and Paul Hanson, University of Wisconsin
• Sensor Systems: By the end of 2004, 6 permanently deployed buoys sampling 6
different lakes, using a suite of commercially available sensors connected
Campbell Scientific data loggers. The buoys are powered by solar panels.
• Communication: Serial Digital Spread-Spectrum (DSS) communication among
the six buoys, which serve a secondary role as the communications backbone of a
larger, more versatile system.
• Data transport: Data transport layer, which relies on vendor-proprietary software
to establish a connection between a fileserver and the buoys over a standard
wireless serial layer, and move the data from the proprietary source (i.e., the buoy)
to a generic data queue. Custom software was written to transport the data to an
ODBC compliant database.
• Information Management: Information Management layer, which uses Web
Services to implement data transport, conduct data QA/QC, and facilitate
communication with additional modeling and data-sharing services.
5 km
Laboratory
Inter net
2 km
Instrumented
Buoy
Serial Spread-
Spectrum Radio
Om ni-directional
Antenna
900 MHz
Radio Signal
Wireless Sensor Network for Acoustic Monitoring
Contact: Stuart Gage, Michigan State University
• Sensor Systems: The system utilizes a small-footprint computer with multiple
data-input capabilities. A Bird Bug® 100M microphone system senses the
acoustic signals, a Creative Audio® sound card digitizes them and High Criteria's
Total Recorder schedules their local storage as digital wav files. Pictures are
acquired through a USB or FireWire camera. A RainWise® MK III Wireless
Weather Station interfaces with the computer via a serial adapter to record wind
speed and direction, temperature, barometric pressure, rainfall and relative
humidity. Communications: A combination of systems are used to transmit data
from remote sites including 802.11, DSL, Broadband, Tx, and satellite (Direct
PC) communications depending of communication availability.
• Data Transport: A series of onboard software programs are used to capture and
store the acoustic and corollary data. High Criteria's Total Recorder uses the
computer's internal clock to trigger 30-second recordings from the microphone at
the predetermined intervals. Acoustic samples are stored as 22.050 kHz 16-bit
monaural wav files on the local hard drive. Webcam 32 synchronizes the visual
snapshots to the acoustic samples and stores the photographs locally as jpegs.
Automize software is used to time FTP transmissions of sound, inages and
weather data to the remote server.
• Information Management: Files are stored on a Snapserver Terabyte (Raid 5)
and are converted to sonograms (Sprectrogram), partitioned using an
Environmental Acoustic Analysis System and processed with the IDRISI image
analysis system as they are received from the field. Oracle is used to archive and
retrieve the results of the analysis and provides access, via interactive web
interface to the sound files and the resulting sonograms.
802.11
b
Micro-comp uter to
sense, digitize and
transmit a digital
audio file (~1.3 Mb
wav) and an image
file (~20 Kb jpg file)
every ½ hour using
FTP over an 802.11b
connection
Local server with
802.11b access
and long range
communication
Web server with access to
digital acoustic and image
library
Clickable ecosystem web site
http://www.cevl.msu.edu
Acoustics Monitoring
in different landscapes
Interactive access to
an Automated
Acoustic
Analysis System for
ecological analysis
and assessment
(Clickable Ecosystem)
Battery
Solar
Panel
Digital
Acoustics
Library
Acoustics Index
Computation
anthrop hon y
bio phony
geophony
Sonogram
Creation
Acoustic signal
partition
(11 levels)
Sonogram
Processing
intensity
diversity
fragmentation
richness
Digital Library
Deposit
Field
Transmitted
Acoustics
Web-Based
Analysis
Compare Locations
Frequency Bands
Monthly Trends
Yearly Trends
Wireless Sensor Network at Virginia Coast Reserve Long Term
Ecological Research site
Contact: John Porter, University of Virginia; Tom Williams Air networking. com
• Sensors Systems: Fixed and pan-tilt-zoom network cameras using Panasonic
security cameras coupled with Axis Communications network video servers and
IQEYE network cameras. Meteorological sensors, and tide pressure sensors are
connected to Campbell Scientific data loggers.
• Communications: The "backbone" network connects the VCR/LTER Laboratory
in Oyster, VA with Broadwater Tower on Hog Island (a relic Navy lookout tower)
22 km away, uses a proprietary WiLAN 900 MHz wireless Ethernet bridge with a
speed of 2 Mbps. Within Hog Island, an 802.11b ("Wi-Fi") wireless Ethernet
cloud over Hog Island and the adjoining lagoon, radiates from two towers, one at
each end of Hog Island. Amplified 802.11b (2.4 GHz) access points and high-
gain omnidirectional antennae provide wireless access to computers and
equipment throughout southern Hog Island at distances up to 10 km (using
directional antennae).
• Data Transport: Vendor-specific proprietary software are used to collect data
from data loggers. Images from web cameras are captured using command-line
web browsers (such as wget from gnu.org) using automatic scheduling software,
or via camera-initiated FTP transfers.
• Information Management: Webcam images are indexed in a MySQL relational
database and can be queried and viewed using web-based visualization and
comparison tools. Data from data loggers is error-checked and summarized using
standard statistical software and the results (both text and graphical) are posted on
the World-Wide Web.
Bird Camera
Meteorological
Station
1.3 km
8 km
Cameras &
Tide Station
22 km
Camera
Laboratory
900 MHz
Wi-Fi Adapter
Directional
Antenna
Wi-Fi Access
Point
Om ni-directional
Antenna
900MHz Wireless
Bridge
Wi-Fi Radio
Signal
Inter net
Broad water
Tower
Machipongo
Station
Wireless Sensor Network at the Santa Marguarita Ecological
Reserve
Contact: Sedra Shapiro and Pablo Bryant, San Diego State University
• Sensor Systems: 802.11b-compatible equipment such as sensors, laptops,
personal digital assistants, and still and video cameras.
• Communications: The network backbone uses the 2.4GHz frequency Ethernet
radios employing 802.11b or VINE protocols. Lucent/Ava ya, Hyperlink and Wi-
Lan radios are used. Access to the network by sensor clients in the field is
accomplished by using 802.11b PC cards in conjunction with serial to TCP/IP
interfaces (e. g. , Wave LAN EC) or equipment that readily supports Wi-Fi using
PC and Compact Flash (CF) cards, such as laptops, Personal Digital Assistants
and digital cameras. Dataloggers that record multiple sensor inputs like the
Campbell Scientifics 23X , link to the network's wireless Access Points using
WaveLANs or they can be directly connected to Ethernet networking switches by
using the Campbell NL100 network interface. Other types radios and dataloggers
can be integrated into our network at the backbone nodes using generic TCP/IP
converters and Ethernet cables.
• Data Transport: Sensor data collected in the field are transmitted to a local FTP
workstation and archived in raw form. Non-proprietary data are simultaneously
forwarded to the Field Station web server on SDSU campus via HPWREN and
the Internet. Individual researchers that are collecting data as part of a grant
project or for a graduate degree and do not want their data freely available to the
public, can use passwords to access their data on the FTP server or have data
automatically FTPed to a secure server of their choice that is accessible by
Internet.
• Information Management: Uses ROADNet Tools that implement a virtual
object ring buffer design to provide a tiered (scalable), distributed, seamless user
interface/data delivery mechanism for data collected in real-time.
Wireless Network of High-Performance Research and Education
Network
Contact: Hans-Werner Braun, Todd Hansen, San Diego Supercomputer Center
• Sensor Systems: IP accessible cameras, and other sensors.
•Communications: High-end radios capable of transporting 45 Mbps of data are
used for the network backbone and for selected links. These include licensed
radios from Stratex and license-exempt 5.8GHz radios. Unfortunately, these high
speed links are expensive. To provide IP connectivity to regional sensor networks,
which have lower data requirements and greater spatial diversity, 2.4 GHz
hardware of various manufactures (i. e., 802.11b compatible or proprietary brands)
are used.
• Data Transport and Information Management : ROADNet, a collaborative
project, provides data transport software that enables recovery in the event of brief
to medium-duration outages. The ROADNet project is developing a near real-time
buffered data transport software package that reliably transports data to end users
while providing a means to do distributed data analysis requests. Together with
the HPWREN project, they are developing an effective real-time data integration
and analysis architecture.
Wireless Network of the Taiwan ECOGRID
Contact: Fang-Pang Lin National Center for High Performance Computing (Taiwan) , and
Chau-Chin Lin, Taiwan Forestry Research Institute
• Sensor Systems: Currently there are three observation sites in Fushan and three in
Nan-Zen-Shan (NZS), and each site have one camera. One infrared camera and
light and one microphone are installed in NZS for capture of both video and
audio. In NZS two observation sites are powered by AC power supply while the
other one is powered by solar panel. The former two sites are connected to a
commercial video server. One buoy is permanently deployed in Yang Yang Lake
(YYL) connected to Campbell Scientific data loggers. Three sites of an underwater
observation system with nine cameras have been established. A commercially
available video server is used to convert the analog image signal to a digital signal
(MPEG1 format) . The zoom lens of cameras could be remotely controlled through
Internet protocol. The whole system is powered by AC 110 volt circuit, as well as
a solar panel.
• Communications: The observation sites are connected via Wi-Fi (IEEE 802.11b)
using Ac-Hoc mode bridges. as well as Serial Digital Spread-Spectrum (DSS)
communication. Five sites built towers to relay the wireless network signal.
• Data Transport: The data transport layer relies on the software provided by
video server vendors. Ecologists can browse the video stream provided by build-
in software of video server, or they can record the video and audio manually with
the stand-alone software provided by video server vendors. The support of SDSC
Storage Resource Broker (SRB) middleware will be added to embedded smart
sensor by the end of 2004. It will help smart sensors to connect to "Grid Resource
Pools". The data transport relies on the ADSL circuits, which connect the LAN of
under water observation systems and the Internet. The data includes live image
data stream and recorded MPEG1 files.
• Information Management: After development, mutual smart sensor, motion
detection and voice analysis capabilities will be added into web service portal. A
smart sensor status monitoring service is in development. Information
management uses Web Services to implement data transport, system diagnosis
and control, and facilitate versatile communication with further modeling and
information sharing services.
pdf