Student Led Test Beds: The CASA Model

1
Student Led Test BedsThe CASA Model

• Jorge M. Trabal, Brian C. Donovan and David J.
  McLaughlin
• NSF ERC 2006 Annual Meeting
• November 30, 2006

2
Jorge M. Trabal

• PhD candidate at the Department of Electrical
  and Computer Engineering at the University of
  Massachusetts
• Future Faculty Member of the Department of
  Electrical and Computer Engineering at the
  University of Puerto Rico, Mayagüez Campus
• B.S. and M.S. from the University of Puerto Rico,
  Mayagüez Campus
• CASA student led test bed leader

3
What is CASA?
CASA stands for Collaborative Adaptive Sensing of
the Atmosphere and is a National Science
Foundation (NSF) Engineering Research Center
(ERC) based in the implementation of Distributed,
Collaborative and Adaptive Sensing (DCAS)
Networks.
Distributed large number of dispersed short
range sensors Collaborative concentrate system
resources in coordinate targeting Adaptive
rapid reconfiguration in response to changing
conditions
NetRad
4
Observation Limitations

• Earth curvature effects prevent 72 of the
  troposphere below 1 km from being observed
• 250 km average spacing between ground radars
• 1-4 km resolution of storms in the upper
  troposphere

5
SLTB Motivation Observation Limitations
Single NEXRAD (TJUA) in PR Distance Cayey to
Mayagüez 100 km Site Elevation 850
m Curvature Beam Elevation (Above Site) 600
m Total Beam Elevation (Above SL) 1.4 km
NEXRAD Rainfall 1 Hour Total
6
Test Bed RD Goals

• Establish a quantitative precipitation estimation
  (QPE) network starting at the western end of the
  island.
• Overcome the NEXRAD coverage gaps caused by
  extended range and terrain blockage.
• Develop a distributive collaborative adaptive
  sensing (DCAS) strategy to improve tropical QPE
  for western Puerto Rico.
• Explore the capabilities of short range, and
  limited node computation Off-the-Grid radar
  network.
• Analyze and target areas more vulnerable to flash
  floods.
• Provide end-user data to support research and
  emergency management process.

10 km Off-the-Grid
30 km Magnetron
PR1 (Stefani) Radar
TJUA NEXRAD
Off-the-Grid Radar
7
Test Bed Educational Goals

• Educate CASA students through practical systems
  engineering and cross-disciplinary collaboration.
• Increase the number of female and
  underrepresented minorities in engineering and
  science fields, and encourage them to pursue
  graduate studies.
• Representation of students in CASA decision
  making and management.

8
SLTB Composition by Field of Study

• 8 different fields of study
• 10 undergraduate students
• 15 graduate students
• - 5 PhD
• - 10 MS

9
SLTB Gender Composition

• 6 female students (32)
• - 1 undergraduate
• - 4 MS
• - 1 PhD
• 19 male students (68 )
• - 9 undergraduates
• - 6 MS
• - 4 PhD

10
SLTB Underrepresented Minorities Representation

• 52 Underrepresented Minorities
• 3 Underrepresented female (12)
• - 1 undergraduate
• - 1 MS
• - 1 PhD
• 10 Underrepresented male (40)
• - 6 undergraduates
• - 3 MS
• - 1 PhD

11
Active Projects
Radars Hardware and Software Design
Puerto Rico Climatology
Antenna Design
OTG Vulnerability Study and Site Evaluation
Grid Based File Storage
Operations Concept
12
System Requirements Document and Design Process

• Core of the system design process
• Serves as a contract between students and
  everyone involved in the development of the
  system
• Increases coordination and organization between
  subprojects
• Facilitates understanding for new SLTB students
• Documented knowledge and innovation

13
Intercampus Communication

• Intercampus system requirements document (SRD)
  and vulnerability assessment biweekly meetings
  are organized
• On-campus weekly or biweekly meetings are
  organized as well
• General students meetings during NSF site visits
  and CASA research retreats
• Intercampus communication media
• - Videoconference calls
• - Teleconference calls
• - Email
• - Skype
• - Instant messaging

14
Barriers to Overcome

• Graduating students leave the test bed team
  almost every year (some project components stall
  temporarily)
• New students coming to continue the work
  (training and learning process time required)
• Deciding on establishing a fixed meeting time is
  a hard task (e.g. time zones, courses, other
  meetings, daylight savings time, etc.)
• Technical language issues between study fields
  (e.g. acronyms, point of view, way of thinking,
  etc.)
• Understanding and using a different language
  (e.g. English for Hispanics)

15
Methods to Overcome the Barriers

• Student pairs working in similar projects or same
  project
• Technology and knowledge transfer through the
  SRD.
• Go through the schedule of all the students and
  find the meeting time that maximizes student
  participation
• Try to prevent the use of acronyms in the
  meetings or if used, explain them prior to use
• During the meetings, translation to Spanish is a
  common practice if required for everyone to
  understand

16
Structure and Schedule
17
Potential Development Opportunities
Colombia
Japan
Sacramento, CA
Vancouver, BC
18
Related Publications
J. M. Santos, J. Miller, A. Rojas, M. Beaton and
X. Ortiz. " Vulnerability to Disasters in Puerto
Rico Incorporating the Social, Physical and
Built Environment to Radar Scanning Strategies,
16th Conference on Applied Climatology, San
Antonio, Texas, January 2007. J. M. Trabal, B.
Donovan, M. Vega, V. Marrero, D. J. McLaughlin
and J. G. Colom. "Puerto Rico Student Test Bed
Applications and System Requirements Document
Development ", ICEE, San Juan, PR, July 2006. B.
Donovan, D. J. McLaughlin, J. Kurose and V.
Chandrasekar. "Principles and Design
Considerations for Short-Range Energy Balanced
Radar Networks" , Proceedings of IGARSS05, Seoul,
Republic of Korea. B. Donovan, F. Junyent,
J. Trabal, J. Torres, D. McLaughlin, and L.
Orama. "A DCAS Network for QPE on the Island of
Puerto Rico" , Proceedings of AIAA Region I
Student Conference, Boston, MA, April 2004. T.
Banka, B. Donovan, V. Chandrasekar, A.
Jayasumana, J. Kurose "Data Transport Challenges
in Emerging High-Bandwidth Real-Time
Collaborative Adaptive Sensing Systems",
Proceedings of INFOCOM2005, Miami, FL, March 2005.
19
Summary

• The Student Led Test Bed is a multi-institutional,
  multidisciplinary team
• Focusing on quantitative precipitation estimation
  at western Puerto Rico while educating students
  during the process
• Great experience for both undergraduates and
  graduates
• Opportunity to increase female and
  underrepresented minorities participation in
  sciences and engineering fields
• Student-directed research influencing overall
  CASA strategy
• Student representation in CASAs executive
  committee
• Number and scope of test bed projects is
  increasing

20
Questions
?