[Firebug-cvs] firebug/web .cvsignore,NONE,1.1 Makefile,NONE,1.1 spie.bib,NONE,1.1 spie_2004.tex,NONE,1.1

[<do...@us...>]

Update of /cvsroot/firebug/firebug/web
In directory sc8-pr-cvs1:/tmp/cvs-serv8574

Added Files:
	.cvsignore Makefile spie.bib spie_2004.tex 
Log Message:
Started an outline for the SPIE paper.

--- NEW FILE: .cvsignore ---
*~ *.dvi *.log *.aux *.bbl *.blg

--- NEW FILE: Makefile ---



spie_2004:
	latex spie_2004
	bibtex spie_2004
	latex spie_2004
	latex spie_2004
	dvips -o spie_2004.ps spie_2004.dvi

--- NEW FILE: spie.bib ---

@InProceedings{woo:a2003,
  author = 	 {A. Woo and T. Tong and D. Culler},
  title = 	 {Taming the underlying hurdles for 
Reliable Multihop Routing in Sensor Networks},
  booktitle = 	 {},
  OPTcrossref =  {},
  OPTkey = 	 {},
  OPTpages = 	 {},
  year = 	 {2003},
  OPTeditor = 	 {},
  OPTvolume = 	 {},
  OPTnumber = 	 {},
  OPTseries = 	 {},
  OPTaddress = 	 {},
  OPTmonth = 	 {},
  OPTorganization = {},
  OPTpublisher = {},
  OPTnote = 	 {},
  OPTannote = 	 {}
}


--- NEW FILE: spie_2004.tex ---
\documentclass[11pt]{article}

\setlength{\oddsidemargin}{0in}
\setlength{\textwidth}{6.5in}
\setlength{\topmargin}{0in}
\setlength{\textheight}{8.5in}

\begin{document}

\title{Sensor network behavior at the urban-rural interface:
making smart structures smarter}
\author{M. M. Chen, A. Sharma, D. M. Doolin\thanks{%
Dept. of Civ. and Env. Eng., UC Berkeley}, S. Glaser, N. Sitar}
\date{\today}
\maketitle 


\begin{abstract}
Abstract here will be a 1 paragraph summary of 
what actually gets written in the paper, and 
should be the very last part of the paper written.
The material in the abstracted submitted in the 
call for papers forms the basis of the introduction.
\end{abstract}

\section{Introduction}

Smart structures incorporating wired and wireless network 
sensors are becoming closer to practical reality.  Interpreting
data from such structures, however, 
requires context.  For example, temperature sensors
fixed to the exterior of a building may record high 
temperatures as a result of radiant heat from a solar
source, or from a wildfire.  Interpreting such data 
requires knowing the environmental conditions
surrounding the structure, which may also be monitored
by networked sensors.  Convenience dictates that 
any such surrounding network be wireless and self-organizing.



Wireless self-organizing sensor networks are the topic of much 
current interest.  One of the most difficult aspects to 
constrain with these networks is the efficacy of radio 
communications as a result of limited power. Most of the ongoing
work in this arena has been built upon testing in relatively 
unobstructed environments, such as flat outdoor areas, and large
indoor spaces. However, more real world environments 
that comprise of man-made of natural obstructions will cause the 
existing architectures -  in the way they organize, share and/or
gather information and do routing, to not perform optimally. 


Preliminary results using 900 MHz radios indicate
that vegetation reduces the reliability of the 
network to transmit data.
Here, we present the results of field experiments using the 
Mica2 platfrom, an  Atmel 128L microprocessor running 
the Tiny Operating System (TinyOS) with 433 MHz radio.  
The network was  deployed in Claremont Canyon, 
a wooded riparian environment 
straddling the Berkeley and Oakland  (CA) 
city limits. Vegetation in Claremont Canyon consists 
of a high proportion of eucalyptus and Monterey pine
groves, extremely flammable in the dry conditions 
common in late summer and early autumn in the area.
The upper ends of the canyon are part of the East Bay 
Regional Parks system, while the lower parts are 
heavily developed with single family residences
and small retail establishments.  


Some of the questions that will need to be answered are: 

\begin{itemize}
\item How ofen does data need to be transmitted? 

\item Determine the signal strength gradient in
  each of the vegetation types. 

\item Does, the position/orientation (in the z direction)
  of the sensor node matter? 
\end{itemize}


\section{Protocols}

\section{TOS active messages}

One paragraph summary of AM here.

\section{Multihop routing}

One paragraph summary of MH6/MH11/BLAST or 
whatever they are calling it when this paper ships.

\subsection{Firebug data protocol}

However many paragraphs it takes to explain our stuff.


\section{Methodology}

This presents an outline of field experiments on a
soft-deployment of a sensor network in the Claremont
Canyon to collect radio communication statistics. 
Most simulation and test environments for sensor
networks are characterized as idealized, ease of line
of sight, smooth terrain.

Testing sensor networks in an outdoor setting has
been attempted mostly in scenarios dealing with
habitat monitoring and form the only precedence for any
field testing. We attempt to use to the MICA2 platform
and BLAST~\cite{woo:a2003} routing and topology algorithm and see how
it performs in a typical grassland environment
with plenty of natural barriers.


The test will consist of 20 nodes spread across various
fuel types in the environment. The metrics for observation
are radio signal strength, and packet loss events (0, 1),
spatial and temporal (long and short term) correlation in 
packet loss, and measuring temperature field gradient.
Volume of data collected should be able to let us draw
conclusions regarding each of these metrics.


\section{Results}


\section{Discussion}

\section{Conclusions}

\bibliography{spie}
\bibliographystyle{plain}

\end{document}