EE290Q: Organization and Management of Ad-hoc Sensor- and Actuator Networks

Spring 2006

We 4:00-6:00pm, 203 McLaughlin

Announcements

PROJECT REPORTING

Each of the two groups should write a report in paper format and submit it to Prof's Rabaey and Wolisz by Wednesday May 17 5pm. Make sure that any problems that you encounter with the test bed are reported as soon as possible. The report should be written in typical paper format (2 columns, minimum font size 10, no more than 6 pages). The report should contain:

Abstract
Description of the problem you are trying to solve and the approach you took
Your results
Description of the challenges and problems you encountered with the test bed
Potential improvements you see
Summary

Week 15

Two lectures:
- We May 3 11am (203 McLaughlin) (Make-up lecture):
  - Ambient intelligence - bringing together sensor networks and multimedia
  - Class wrap-up
- We May 3 4pm (203 McLaughlin)
  - Automated synthesis of sensor network applications to networks
  - In network processing (continued)
Reading assignment by Chris Baker.

Week 14

Lecture on distributed processing in sensor networks (Prof. K. Ramchandran)
Reading assignment by Jose Ortiz.
Make up lecture for Week 13 on We May 3 at 11am in 203 McLaughlin (this will be the future perspective class with remote participation of Prof. Wolisz). Final lecture of the class will be held at 4pm the same day.

Week 13

No Lecture - Make up Lecture in Week 15

Week 12

Lecture by Kris Pister on the management and deployment of sensor networks - an industrial perspective.
Reading assignment by Daekyeong Moon
NO LECTURE NEXT WEEK. Make up lecture potentially scheduled for May 3 (4 hours in a day, indeed)
Phase 1 of project to be completed by We April 12:
The project should aim at using ONE OF THE possible methods for localization of the position of
OTHER nodes, using the position of the anchor nodes as reference. You can select ANY of the localization methods to solve the problem. You are expected
a/ to select a method
b/ to implement it
c/ to deliver YOUR estimate of the position of the Non-anchor nodes.
At the end the correct result will be published (we KNOW the real positions!)
and we will see who had the best estimates.

By Wednesday April 12, as a first step you are expected to submit to us:
- the composition of the group
- description of the selected method (step a/ from above)

Week 11

Reading assignments

Week 10

Reading assignment this week: Jana Van Gruenen

Week 9

Reading assignments this week: Delynn Bettencourt

Week 8

Reading assignments this week: Jose Ortiz, Omar Bakr
PROJECT INFO: Access to the testbed, check slides for lecture 8.

Week 7

Reading assignments this week: Dima Ryazanov, Ian Tan, Daekyeong Moon
PROJECT INFO: For more information on TinyOS and the Berlin Testbed, please check this presentation.

Week 6

Reading assignments this week: Xiaofan (Fred) Jiang.
Project will be launched next week - more info in class today.
Adam Wolisz starting to lecture this week.

Week 5

Reading assignments this week: Delynn Betttencourt and Seung-Bum Suh.

Week 4

Assignments this week: Seung-Bum Suh and Omar Bakr.
Reading assignment presentations start this week. Look here for the final schedule.
Project will be officially launched in 2 weeks.

Week 3

URGENT - sign up for reading assignments! Look here for the current schedule.

Week 2

Sign up for Reading Assignments by Tu Jan. 31. Select 2 topics and corresponding weeks (see class schedule) and mail to jan@eecs.berkeley.edu. First come, first serve!

Instructors

Prof. Jan M. Rabaey, 511 Cory Hall, 666-3102, jan@eecs.berkeley.edu
Office hours: Mo 3:30-5pm, 511 Cory Hall.

Prof. Adam Wolisz (Prof. TU Berlin, Adjunct Prof. UCB), 511 Cory Hall, 666-3102, wolisz@eecs.berkeley.edu
Office hours: TBD.

Administrative Assistant

Jessica Budgin, 558 Cory Hall, 643-0694, jessica@bwrc.eecs.berkeley.edu

Mail

To send broadcast messages to the complete class, use ee90q-students@bwrc.eecs.berkeley.edu

Course Description

Wireless sensor and actuator networks are rapidly gaining major traction in a wide range of application areas. To be successful in the commercial arena however, a number of important criteria have to be met. First, it is essential that the individual transceiver nodes are tiny, easily integratable into the environment, and have negligible cost. Most importantly, the nodes must be self-contained in terms of energy via a one-time battery charge or a replenishable supply of energy scavenged from the environment. Realizing these very low power levels requires a vertical system-level design approach, engaging all levels of the design abstraction (from aggressive new circuit approaches over innovative networking and distributed computing techniques). Unfortunately, getting to the cost, size and power numbers needed for a truly ubiquitous deployment, comes with a penalty in reliability. Rather than falling back on traditional reliability enhancing techniques that compromise the energy-efficiency and cost of the individual nodes, a more effective solution is to rely on the unique nature of these networks, that is the ubiquitous availability of nodes. Doing so requires crisp and clearly defined abstraction layers. Another challenge that is often overlooked is the ease of deployment, configuration and management of the network. Again, it can argued to well defined abstraction layers go a long way in making this possible.

In this seminar series, we will traverse the wireless sensor and actuator paradigm in a bottom-up fashion. Starting from implementation constraints and properties of the wireless medium, we will explore the trade-off's at the all layers of the abstraction hierarchy up to the application layer. Metrics such as energy efficiency, robustness and ease of deployment will carry prominently throughout the semester. Real-life case studies will be used extensively.

Student participation and discussion will be an essential part of the course.

Course Links

Comments and questions should be addressed to ee290q@bwrc.eecs.berkeley.edu