04-800-C-Carnegie Mellon University Africa - Carnegie Mellon University

04-800-C

04-800/C
Special Topics in ICT: Geosensor Networks

Core/Elective:  Elective
Units: 12
Lecture/Lab/Rep hours/week: 
3 Lecture Hours Per Week
4 Lab Hours Per Week
Semester/year offered (fall/spring, even/odd/all years): fall
Pre-requisites:  some programming skills

Course Website

sites.google.com/site/geosensornet/

Course description: 

The course introduces into a topic of rising importance for many domains: the collection and processing of spatial data collected from a variety of sources. “Spatial is special” is one of the credos of geography and its modern descendent spatial information theory. Today’s data analytics approaches take parameters of context, such as space and time, as fundamental means to pre-structure large databases following techniques from the domain of geographical information systems research. Moreover, novel, distributed local processing techniques, called spatial computing methods are emerging. Information collected by a sensor network that is processed locally can avoid many of the legal and ethical issues raised by large scale data collection. 
This course introduces into the fundamental technologies employed for spatial computing. Fundamental technologies for developing geosensor network applications are explained: wireless sensor networks, geographic information systems, spatial reasoning, and distributed algorithms. Students learn the fundamental concepts and theories, and apply these in practical lab assignments on four project assignments. The theoretical understanding for the four topics is assessed in two exams.
  

Learning objectives:

The aim of this course is to give students an insight into an exciting emerging field of research and technology. Theory and practice of the underlying technologies are combined tightly, so as to give students both the ability to apply these technologies practically, as well as to do research in the area. Open research topics and the challenges for current approaches in spatial computing, such as the lack of reliable top-down analysis tools, or the trade-offs in geosensor network design, will be discussed and highlighted in each of the lab assignment projects on the basis of original research articles.

Outcomes:

After completing this course, students should be able to: 
Load and visualize spatial data using a GIS (GRASS GIS, and potentially ArcGIS)
Understand the specifics of spatial databases
Use spatial analysis tools to identify and track, e.g., hotspots and events
Develop WSN applications with ZigBee nodes and simpler nodes
Develop and test distributed reasoning algorithms
Understand requirements for sensing applications for the geographic scale

Content details:

1. Introduction: wireless sensor networks, geographical information systems, spatial data formats, spatial representation, crowd-sourced data collection
2. Wireless sensor networks fundamentals: foundations, sensors and sensing, ZigBee nodes, energy saving techniques, energy harvesting
3. Aggregating and analyzing information from sensor networks and crowd-sensing
4. Spatial information theory: models of space, geometric and topologic models, spatio-temporal models
5. Distributed spatial reasoning: spatial reasoning formalisms, localized processing of spatial reasoning mechanisms
6. Monitoring spatial change: distributed spatio-temporal reasoning formalisms

Literature:

Matt Duckham: Decentralized Spatial Computing - Foundations of Geosensor Networks. Springer 2013

Students’ assessment:

Four assignments, midterm exam, final exam