Howdy All,
Here are a few good reads that can give you a better feel of sensor networks. Most of them can be found directly through citeseer or google. Three that I recommend: "Next Century Challenges: Scalable Coordination in Sensor Networks", "System architecture directions for network sensors", and "The SOS paper".
Note that most of these papers are a little bit dated. I am of the opinion that it is better to get a solid foundation, before chasing after new ideas that may flop. I have tried to give you some information on why the paper is worth reading, some fun / interesting background, and related areas if you find it interesting:
================================
- Why: One of the first papers to talk about sensor networks and in network aggregation. Foundation paper.
- Fun Side Note: Just this past year, people (including the authors) are beginning to question if in network aggregation is the right way to go.
- Related Papers: Anything on "directed diffusion".
Next Century Challenges: Scalable Coordination in Sensor Networks. Deborah Estrin , John Heidemann , Ramesh Govindan and Satish Kumar In Proceedings of the Fifth Annual International Conference on Mobile Computing and Networks (MobiCOM '99), August 1999, Seattle, Washington.
http://www.google.com/url?sa=U&start=1&q=http://www.isi.edu/~johnh/PAPERS/Estrin99e.html&e=10187
http://www.isi.edu/~johnh/PAPERS/Estrin99e.pdf
================================
================================
- Why: Describes one of the original wireless sensor network nodes (closely related to the Mica2 that you are using) and the first big OS for sensor networks called TinyOS. SOS (the operating system that you are using) is heavily influenced by TinyOS and a lot of the world uses TinyOS.
- Fun Side Note: This collection of authors hold a lot of sway in the sensor network community. I may not always agree with what they say (although I often do), but I always listen closely.
- Related Papers: Anything on "tinyos".
System architecture directions for network sensors. Jason Hill, Robert Szewczyk, Alec Woo, Seth Hollar, David Culler, Kristofer Pister . ASPLOS 2000.
http://www.tinyos.net/papers/tos.pdf
================================
================================
- Why: This is one of the real deployments that has been used to study a habitat. We will probably be aiming at a scaled down version of this for the summer.
- Fun Side Note: This deployment showed the sensor network community how hard deployments really are. Batteries died, sensors rusted, software crashed. It is trying to handle all of these that makes sensor networking so much fun.
Wireless sensor networks for habitat monitoring. Alan Mainwaring, Intel Research Laboratory; Joseph Polastre, Robert Szewczyk and David Culler, University of California, Berkeley.
================================
================================
- Why: A lot of the early work in sensor networks used simple uniform circular radio modules to simulate radio range (ie, if you are within x meters of the node you can hear it 100%, else you can not hear it at all). This resulted in a lot of protocol development and tuning that is worthless. In reality a simple uniform disk radio model is bogus and the work based on that assumption was of little real use.
- Fun Side Note: Check out your assumptions before basing a lot of work on them!
- Related Papers: The Mistaken axioms of wireless-network research, D. Kotz, C. Newport, C. Elliott, Dartmouth College Computer Science Technical Report, TR2003-67 (I have not read it but I should).
http://www.cs.dartmouth.edu/~dfk/papers/kotz:axioms-tr.pdf
SCALE: A tool for Simple Connectivity Assessment in Lossy Environments. Alberto Cerpa, Naim Busek, and Deborah Estrin CENS Technical Report 0021, September 5, 2003.
================================
================================
- Why: MAC layers are very important for energy efficient sensor network operation. This is one generation back of what is being used on high end systems. It shows the complexity that comes out in the MAC layer.
- Fun Side Note: Writing a MAC layer is apparently horribly painful. Even with spectrum analyzers, there is no good way to debug them.
An energy-efficient MAC protocol for wireless sensor networks. Wei Ye, John Heidemann and Deborah Estrin. IEEE Infocom New York, NY, USA, June, 2002.
http://www.isi.edu/~johnh/PAPERS/Ye02a.pdf
================================
================================
- Why: Time synchronization (like the MAC) is very important for low energy operation. The motivation is to have all nodes go to sleep (low power), and then wake up at the same time to send data.
- Fun Side Note: Umm.. I can't think of anything.
Timing-sync Protocol for Sensor Networks, Saurabh Ganeriwal, Ram Kumar, Sachin Adlakha and Mani Srivastava, ACM Sensys 2003.
================================
================================
- Why: This is an example of the fun things that we can start to do when larger sensor networks are deployed. Edges are pretty cool.
- Fun Side Note: Okay. I need to get back to research and stop writing "fun side notes".
Localized Edge Detection in Sensor Fields by Krishna Kant Chintalapudi, Ramesh Govindan.
http://www.google.com/url?sa=U&start=1&q=http://cs.usc.edu/~ramesh/papers/edge_detection.pdf&e=10187
================================
================================
- Why: Ends up that virtual machines provide a lot of advantages in sensor networks. Among other things, it makes it very easy to retask parts of your system.
- Fun Side Note: One guy from Colorado managed to get a Python VM running on the Mica2 mote. Ahh, yeah!
- Related Work: Any thing on "sensor network virtual machine".
P. Levis and D. Culler, Maté: a Virtual Machine for Tiny Networked Sensors , ASPLOS, Dec 2002.
http://www.google.com/url?sa=U&start=1&q=http://www.cs.berkeley.edu/~pal/pubs/mate.pdf&e=10187
================================
================================
- Why: This is the operating system that you are spending the summer with. Probably worth understanding some of the internals.
- Fun Side Note: SOS is starting to be used by the East cost sensor networking community on higher class systems running 32-bit (rather than 8-bit) processors. Go figure.
This paper will soon be published in Mobisys.
"A Dynamic Operating System for Sensor Nodes" by Chih-Chieh Han, Ram Kumar, Roy Shea, Eddie Kohler and Mani Srivastava
University of California, Los Angeles
================================
Finally, if you are looking for more foundation papers you can check out:
http://www.cens.ucla.edu/CS213/
for a pretty complete list of papers up through 2003. More recent papers will need to be found through the sensor network journals.