Wireless sensor Grid - Reports LTER ASM Meeting, Workshop on Sensor Networks; NSF Workshop Report on Environmental Cyberinfrastructure Needs for Distributed Sensor

1
Wireless sensor Grid - ReportsLTER ASM Meeting,
Workshop on Sensor Networks NSF Workshop
Report on Environmental Cyberinfrastructure Needs
for Distributed Sensor

• Wireless Sensor Networks and Their Applications
  in the Environment
• Thursday 29 January 2004
• Peter Arzberger

2
Long Term Ecological Research Network

• LTER Network is a collaborative effort
• More than 1100 scientists and students involved
  investigating
• Ecological processes over long temporal and broad
  spatial scales.
• The Network promotes synthesis and comparative
  research across sites and ecosystems and among
  other related national and international research
  programs. 
• The NSF established the LTER program in 1980 to
• Support research on long-term ecological
  phenomena in the United States.
• Provide information for the identification and
  solution of ecological problems
• The 24 LTER Sites represent diverse ecosystems
  and research emphases
• The LTER Network Office coordinates
  communication, network publications, and
  research-planning activities.

http//www.lternet.edu
3
Long Term Ecological Research Network
1. Andrews LTER (AND)2. Arctic LTER (ARC) 3.
Baltimore Ecosystem Study (BES) 4. Bonanza Creek
LTER (BNZ)  5. Central Arizona - Phoenix
(CAP)  6. Cedar Creek LTER (CDR) 7. Coweeta LTER
(CWT) 8. Harvard Forest (HFR) 9. Hubbard Brook
LTER (HBR)  10.Jornada Basin (JRN) 11.Kellogg
Biological Station (KBS) 12.Konza LTER
(KNZ) 13.Luquillo LTER (LUQ) 14.McMurdo Dry
Valleys (MCM) 15.Niwot Ridge LTER (NWT) 16.North
Temperate Lakes (NTL) 17.Palmer Station
(PAL)  18.Plum Island Ecosystem
(PIE)  19.Sevilleta LTER (SEV) 20.Shortgrass
Steppe (SGS) 21.Virginia Coast Reserve
(VCR) 22.Florida Coastal Everglades
(FCE) 23.Georgia Coastal Ecosystems
(GCE) 24.Santa Barbara Coastal (SBC)
4
International LTER Network
Launched in 1993 http//www.ilternet.edu
Current Chair, ILTER Hen-Biau King, TFRI
5
Exploring New Spatial and Temporal Scales in
Ecology Using Wireless Sensor Networks

• September 2003 All Scientist Meeting of the Long
  Term Ecological Research
• Participants
• Tim Kratz, Paul Hanson North Temperate Lakes
• Stuart Gage Kellogg Biological Field Station
• Hen-biau King, TERN and Fang-Pang Lin, NCHC
• John Porter Virginia Coast Region
• Bill Michener LTER Network Office

6
Goals of LTER Workshop

• To identify scientific research opportunities and
  areas enabled and opened up by wireless sensor
  networks
• New Science
• Cross-Site or Synthetic Research
• Impact of working at new spatial or temporal
  scales
• To exchange information on capabilities,
  techniques and technologies, and experiences for
  wireless sensor networks
• Lessons Learned
• Biggest Challenges
• Develop products that help achieve the goals above

7
VCR/LTER Wireless NetJohn Porter, Tom Williams,
Dave Smith

• The VCR/LTER uses a hybrid network with both
  proprietary 900 MHz and standard WiFi 802.11b 2.4
  GHz wireless Ethernet connections.
• Areas within line of sight of our two towers are
  tinted in yellow

http//www.lternet.edu/sites/vcr/
Source John Porter, Virginia Coast Reserve
8
Uses of Wireless at VCR/LTER
Integrated camera/ web server/radio/power

• Real-time Meteorological Tide data
• Web Cameras (6 currently deployed)
• Access to networked data resources (e.g., the
  web) in the field

Source John Porter, Virginia Coast Reserve
9
Uses of Webcams

• Capture time series
• Education
• Non-obtrusive observation
• Observe rare events

A picture is worth a thousand words
Source John Porter, Virginia Coast Reserve
10
Wireless Webcam pre Isabel

• Source John Porter, Virginia Coast Reserve

11
During Isabel
Source John Porter
Source John Porter, Virginia Coast Reserve
12
Early Isabel
Source John Porter
Source John Porter, Virginia Coast Reserve
13
Peak Flooding
Source John Porter, Virginia Coast Reserve
14
Isabel Winds
Sensors can be where it is too dangerous for
humans
Source John Porter, Virginia Coast Reserve
15
Some lessons learned

• Power supplies, not radios, are the most
  difficult component
• Most consumer-grade DC-DC voltage converters are
  power hogs
• Use cheap inverters, not expensive ones
• The cheap ones reset automatically if batteries
  are drawn down, expensive ones dont.
• Use digital, not analog timers to cut down on
  hours of operation to save power
• Cheap inverters have poor frequency control

Source John Porter VCR
16
North Temperate Lakes
Source Paul Hanson, NTL
Freshwater important for human survival habitat
important of other species
http//www.lternet.edu/sites/ntl/
17
North Temperate Lakes University of Wisconsin
Automated Sampling Buoys
Source Paul Hanson, Tim Kratz, NTL
Sensors
Picture of Lab Freewave
Picture of Buoy Freewave
Communication
18
Continuous monitoring provides opportunity for
pattern discovery And understanding relationships
between variable
Source Paul Hanson, Tim Kratz, NTL
19

• Successes (the dedicated network)
• Building large platforms
• Establishing bi-directional communication
• Publishing data to the Web
• Studying processes contained within an ecosystem
• Studying processes at few spatio-temporal
  resolutions

• Current Exploration (the adaptive network)
• Optimizing wireless networks
• Auto-configuring ad hoc networks
• Managing the data load from ad hoc networks
• Managing power
• Distributing data to diverse clients
• Developing network intelligence
• Studying links across ecosystem boundaries
• Studying processes at multiple spatio-temporal
  resolutions

Source Paul Hanson, Tim Kratz, NTL
20
Where to from here?
Better power sources More radio
range Communication among sensors Adaptive
Sampling run by intelligent agents Scalable
systems
Source Paul Hanson, Tim Kratz, NTL
21
Development of Wireless Instrumentation for
Remote Environmental Acoustic Sensing
Stuart Gage Computational Ecology and
Visualization Laboratory Michigan State University
http//www.lternet.edu/sites/kbs/
Source Stuart Gage, KBS
22
Sound as an Ecological Indicator and a Stressor
As an Ecological Indicator- The integrity and
dynamics of an ecosystem may be correlated to the
complexity of that ecosystems soundscape.
As a Stressor- Organisms require communication
for their survival. Organism population may be
inversely proportional to the degree of acoustic
disruption.
Source Stuart Gage, KBS
23
Environmental Acoustic Monitoring Infrastructure
Source Stuart Gage, KBS
24
EcoGrid Expanding
Fushan
http//ecogrid.nchc.org.tw/
Source Fang-Pang Lin
25
HPWRENconnected topologyagendaMay 2002
Santa Margarita Ecological Reserve
Palomar Observatory
Los Coyotes Indian Res.
Pala Indian Res.
Pauma Indian Res.
Rincon Indian Res.
La Jolla Indian Res.
Mesa Grande Indian Res.
San Pasqual Indian Res.
Santa Ysabel Indian Res.
Mt. Laguna Observatory
UCSD/SDSC
SIO
Scripps Pier
Backbone/relay node Science site Researcher
location Education site Incident mgmt. site
http//hpwren.ucsd.edu/
Courtesy Hans-Werner Braun
26
Mt. Woodson area
to UCSD
to North Peak
to Indian Reservations
to Dan Cayan
Doug Bartlett
Hans-Werner Braun
Courtesy Hans-Werner Braun
27
HPWREN Applications

• Ecology
• Stream Sensors,
• Behavioral Ecology
• Oceanography
• Astronomy
• Earthquake Engineering
• Geophysics
• Crisis Management
• Distance Education

Multiple applications on same wireless backbone
28
Instrumenting the Environment
Courtesy NSF Brochure
29
This model can be replicated and scaled to meet
the challenges of global environmental observing,
analysis, and action
http//www.nsf.gov/pubsys/ods/ getpub.cfm?nsf04549
30
ParticipantsDeborah Estrin, Bill Michener,Greg
Bonito Total more than 85 AP Community
Masayuki Hirafuji (NARO), Fang-Pang Lin
(NCHC), Shinji Shimojo (Osaka)
http//lternet.edu/ sensor_report/ cyberRforWeb.pd
f
31
Overarching View Sensor Networks

• Revolutionary Tool for Studying the Environment
• Enables Scientists to Reveal Previously
  Unobservable Phenomena
• New Cyberinfrastructure Capabilities and
  Infrastructure, Methodology, Middleware, People
  Needed
• These will lead to paradigm shift in science

32
Overarching View Sensor Networks

• Revolutionary Tool for Studying the Environment
• Spatially extended networks of multivariable
  intelligent sensor arrays are seen as
  revolutionary tools for studying the environment.
  
• Enables Scientists to Reveal Previously
  Unobservable Phenomena
• The temporally and spatially dense monitoring
  afforded by this technology portends a major
  paradigm
• New Cyberinfrastructure Capabilities and
  Infrastructure, Methodology, Middleware, People
  Needed
• To realize this vision will require above and a
  community of multidisciplinary scientists and
  engineers
• To pose newly-enabled scientific questions.

33
Vision of Environmental Sensor Networks

• SCALE Pervasive in situ sensing of the broad
  array of environmental and ecological phenomena
  across a wide range of spatial and temporal
  scales.
• INFRASTRUCTURE Sensor networks should be robust
  and autonomous, be inexpensive and long-lived,
  have minimal infrastructure requirements, and be
  flexible (expandable and programmable) and easily
  deployed and managed
• DATA Sensor network data should be maximally
  self-documenting and of known quality, readily
  integrated with other sensor data, and easily
  assimilated.

34
Key Areas of Discussion and Recommendations

• Sensing Technology
• Deployed Sensor Arrays
• Cyberinfrastructure for Sensor Networks
• Error Resiliency
• Security
• Data Management
• Metadata
• Analysis and visualization
• Education
• Outreach
• Collaboration and Partnerships

35
Sensing Technology

• What are the greatest needs for sensor component
  development for the different communities
  represented?
• Recommendations
• Design more capable sensors
• Long-term integrity
• Performance
• Interactivity
• Minimal environmental impact
• Minimal Power consumption

36
Deployed Sensor Arrays

• What are the most urgent needs in relation to
  deploying sensor arrays in the field to achieve
  the overarching vision of the report?
• Recommendations
• Invest in prototyping and end-to-end testbeds
• Tested in large-scale natural environments across
  range of applications
• Validation, comparison with traditional
  monitoring systems
• Sensor networks include sensors, network
  security, information technologies
• Automated system layout and coverage estimation
  composition and configuration of synthetic and
  simple sesnors validation and calibration of
  sensor systems

37
Cyberinfrastructure for Sensor Network

• Support new genre of cyberinfrastructure research
  and development for scalable sensor arrays
• Middleware and services (time synchronization,
  localization, in situ calibration, adaptive duty
  cycling, programmable tasking, triggered imagine)
  needed for hyper-scalability, sustainability, and
  heterogeneity
• Build the requisite Grid and Web services
• To convert raw environmental data into
  information and knowledge

38
Security and Error Resiliency

• How can we construct flexible, light-weight
  systems that are secure? (e.g. not excessively
  vulnerable to denial of service, inappropriate
  access)
• How do we best characterize and optimize data
  quality from systems composed of large numbers of
  noisy/faulty channels?
• Recommendations
• Need solutions for free/open access to most data,
  but protect network, sensors, and sensitive data
• Need tools for self-diagnosis and self-healing of
  network, and resilient operation when some nodes
  compromised.

39
Metadata and Data Management

• What metadata developments are needed to promote
  data discover, access, integration, synthesis?
• How do we best manage diverse, heterogeneous data
  streams (biological, physical, chemical )?
• Recommendations
• Support development of metadata tools (for
  automated metadata and data encoding)
• Engage community in standardization efforts
• Community includes sensor developers, users,
  informatics specialists, standards organization
• Focus on design, development implementation,
  testing, adoption stages

40
Analysis and Visualization

• What tools are needed for analyzing and
  visualizing complex, multidisciplinary, spatially
  extended data?
• Recommendations
• Algorithm development drawing from statistics,
  machine learning, visualization
• Work in high-bandwidth sensor streams
• Tools for mobile devises
• Tools that integrate high-resolution imagery and
  video, high-fidelity audio and tactile interfaces
  to support virtual and augmented reality
  environments

41
Education and Outreach

• Train in interdisciplinary setting
• Outreach to public, decision-makers and resource
  managers needed
• Information systems needed

42
Collaborating and Partnering

• Build Partnerships
• Universities, research labs, industry, standards
  organizations
• Sustain long term deployment
• To keep facilities alive, evolving, and
  non-obsolescent
• Need funding for staffing for stewardship and
  management
• Promote open source solutions and repositories
• Need incentives for and ease of contributing to
  open source toolsets, models and testbeds
• Allow for developing reusable system components
  and enhancing interoperability

43
Examples in Report

• CUAHSI Consortium of Universities for the
  Advancement of Hydrologic Science, Inc.
• GEON the Geosciences Network
• SpecNet Spectral Network
• Embedded Networked Sensing
• NSF CLEANER Initiative Collaborative
  Large-scale Engineering Analysis Network for
  Environmental Research
• Fixed Ocean Observatories (Neptune)
• NEON National Ecological Observatory Network
• North Temperate Lakes Monitoring
• Observing the Acoustic Landscape (KBS)

44
Role for APAN in Sensor NetworksSome Thoughts
for Discussion

• Forum for discussion
• Topics
• Sensor technology
• Grid and web services
• Networking needs
• Application drivers
• Communities
• Grid working group and current/ future partners
  ApGrid, PRAGMA,
• Natural Resources working group and
  current/future partners
• Networking working with partners
• Catalyst for testing sensor nets
• Place where new technologies are tested in a
  diverse set of environmental conditions

45
Reports

• Exploring New Spatial and Temporal Scales in
  Ecology Using Wireless Sensor Networks, September
  2003 All Scientist Meeting of the Long Term
  Ecological Research http//atlantic.evsc.virginia.
  edu/jhp7e/wireless/
• Environmental Cyberinfrastructure Needs for
  Distributed Sensor Networks, D. Estrin, W.
  Michener, G. Bonito, August 2003 NSF Workshop,
  http//lternet.edu/sensor_report/cyberRforWeb.pdf
• Scalable Information Networks for the
  Environment, A.Withey, W.Michener,
  P.Tooby,October 2001 NSF Workshop,
  http//www.sdsc.edu/pbi/sine_report_pt1.PDF

46
Centers

• Center for Embedded Networked Sensing PI Deborah
  Estrin, http//www.cens.ucla.edu/
• California Institute of Telecommunications and
  Information Technology, http//www.calit2.net

47
Projects

• High-Performance Wireless Research and Education
  Network (HPWREN), PI. Hans-Werner Braun (NSF),
  http//hpwren.ucsd.edu
• Real-time Observatories, Applications and Data
  management Network (ROADNet), PI John Orcutt
  (NSF) http//roadnet.ucsd.edu/
• Bringing the information superhighway to the
  dirt road and the high seas
• National Ecological Observatory Network,
  http//www.nsf.gov/bio/neon/start.htm
• Infrastructure for Biology at Regional to
  Continental Scales (IBRCS), A community resource
  by AIBS, http//ibrcs.aibs.org/core/index.asp