Changeset 2792

Timestamp:

10/30/09 17:05:42 (4 weeks ago)

Author:

jaein

Message:

 

Location:

HydroWatch/Tim/doc/ipsn10

Files:

• sec_adaptive.tex (modified) (2 diffs)
• sec_eval.tex (modified) (3 diffs)

HydroWatch/Tim/doc/ipsn10/sec_adaptive.tex

@@ -2 +2 @@
 \section{Adaptive Scheduling for Energy Management}\label{sec:adaptive}
 
-\subsection{Linking Utility and User Policy}~\label{sec:schedule}
+\subsection{Linking Utility and User Policy}~\label{sec:utility_policy}
 As discussed in Section~\ref{sec:motivation}, one of the central aims of this paper is to argue to we should move beyond an ``always on'' model for environmental sensing. Whilst it is important that nodes in the network have some level of responsiveness (e.g. for reporting data or receiving user commands), by removing the cost burden brought about by continuous responsiveness, we can move energy resources onto other roles. Given this revised model of thinking, we argue that a useful node and network level \emph{utility} can be defined as a combination of metrics for both \emph{network responsiveness} and \emph{data fidelity}. Furthermore we argue that a suitable utility function (which is by definition somewhat subjective) can be inferred from a combination of these parameters and a \emph{user-policy} which defines bounds for network performance.
 
@@ -151 +151 @@
 where $E_{sleep}$ is the total energy consumption of the node in sleep state, $E_{samp}$ is the consumption in the sample state, $E_{lr}$ is consumption for sending low-res reports, $E_{hr}$ is the consumption for streaming full fidelity data samples from storage and $E_f$ is the forwarding cost for passing on both low-res and high-res samples.
 
-Given the parameters of sample frequency and low-res report frequency $(F_{s}(n,k)$, $F_{r}(n,k))$, as described in Section~\ref{sec:schedule}, we can define the energy consumption, in a given interval, as a function of these parameters as: 
+Given the parameters of sample frequency and low-res report frequency $(F_{s}(n,k)$, $F_{r}(n,k))$, as described in Section~\ref{sec:utility_policy}, we can define the energy consumption, in a given interval, as a function of these parameters as: 
  \begin{align}\label{equ:Econsump}
         E_{samp}(n,k) &= A_1(n) F_s(n,k) \nonumber \\ 

HydroWatch/Tim/doc/ipsn10/sec_eval.tex

@@ -51 +51 @@
 %\end{figure}
 
-\begin{figure}[ht]\label{fig:util_cdf1}
+\begin{figure}[ht]
 \centering
   \includegraphics[width=8cm]{fig/util_cdf1}
@@ -58 +58 @@
 \end{figure}
 
-\begin{figure}[ht]\label{fig:UvsEh1}
+\begin{figure}[ht]
 \centering
   \includegraphics[width=8cm]{fig/UvsEh1}
@@ -82 +82 @@
 
 Figure~\ref{fig:multi-cdf1} shows the effect of multi-hop in the case of a binary network tree. The use of $\gamma(n)$ reflects the current operating preference where nodes closer to the leaf pull back in their sampling, (thus reduced utility) in order to lessen the forwarding load on nodes closer to the sink.
-\begin{figure}[ht]\label{fig:multi-cdf1}
+\begin{figure}[ht]
 \centering
   \includegraphics[width=8cm]{fig/multi-cdf1}