CS795/895 - Adventures in Sensor Networks

Fall 2012

Instructor: Prof. Stephan Olariu
Phone: 683-3915

Course Description: A wireless sensor network consists of a large number of cooperating sensors. Interactions among individual sensors and the behavior of the sensor network as a whole are subject to numerous constraints and limitations due to inherent properties of sensor network technology including node capabilities and form factor.

Sensor network applications are wide-ranging and include military sensing, physical security, air traffic control, surveillance, industrial and manufacturing automation, distributed robotics, environment and structure monitoring, and medical sensing, along with many others. Sensors can be deployed or embedded almost anywhere in the physical world under a broad spectrum of conditions from the regular to the extremely harsh. Optical sensors, for example, can be deployed inside automobile engines. These sensors, augmented with computing and communication capabilities can self-organize into a sensor network to achieve higher level coordinated behavior.

As the title indicates, the class is intended to be heavily slanted towards programming the sensors in our lab. Since these sensors are tiny devices with a "mind of their own" the participants should expect to spend a substantial amount of time learning how to program the sensors to accomplish a number of non-trivial tasks.

Topics to be covered: Roughly the first half of the semester will be lecture-oriented with the instructor reviewing a number of topics related to sensor networks. The second half, will consist almost exclusively of lab sessions. The topics to be covered include:

• A review of wireless communications
• A review of wireless network protocols
• Introduction to sensor networks
• Fundamental protocols for sensor networks
• Retasking sensor networks
• Applications

Prerequisites: graduate standing in Computer Science, Computer Engineering, Mathematics or Electrical Engineering. Previous exposure to networks is highly desirable but not essential. Good programming skills and an ability to deal with small devices are a must.

Text: no formal text; a number of relevant papers from recent journal publications and conference proceedings will be discussed in class.

Grading Scheme:

• Class presentation 30%
• Class participation 20%
• Programming project 50%

Office Hours: to be arranged

Back to Dr. Olariu's Home Page