bug 1362: paper update

doc/papers/2011/europar/dispersed-signal.jgr

@@ -15,6 +15,14 @@ newgraph
     size 2
     shell : seq 0 3 | awk '{ f = (512 + 200 + $1 * 1/16/3)*200/1024; printf "hash_label at %d : %.3f\n",$1,f; }'
 
+newline
+  linetype dotted
+  pts 0.82 -1 0.82 4
+(*
+copycurve
+  pts 0.52 -1 0.52 4
+*)
+
 newline
   color 0 0 1
   linethickness 2.0
@@ -22,24 +30,15 @@ newline
   pts
     shell : ./dispersed-signal-data-2.sh 0
 
-newline
-  color 0 0 1
-  linethickness 2.0
-  linetype solid
+copycurve
   pts
     shell : ./dispersed-signal-data-2.sh 1
 
-newline
-  color 0 0 1
-  linethickness 2.0
-  linetype solid
+copycurve
   pts
     shell : ./dispersed-signal-data-2.sh 2
 
-newline
-  color 0 0 1
-  linethickness 2.0
-  linetype solid
+copycurve
   pts
     shell : ./dispersed-signal-data-2.sh 3
 
@@ -63,6 +62,10 @@ newgraph
     shell : seq 0 3 | awk '{ f = (512 + 200 + $1 * 1/16/3)*200/1024; printf "hash_label at %d : %.3f\n",$1,f; }'
     nodraw
 
+newline
+  linetype dotted
+  pts 0.82 -1 0.82 4
+
 newline
   color 1 0 0
   linethickness 2.0
@@ -70,24 +73,15 @@ newline
   pts
     shell : ./dispersed-signal-data-2.sh 0 noshift
 
-newline
-  color 1 0 0
-  linethickness 2.0
-  linetype solid
+copycurve
   pts
     shell : ./dispersed-signal-data-2.sh 1 noshift
 
-newline
-  color 1 0 0
-  linethickness 2.0
-  linetype solid
+copycurve
   pts
     shell : ./dispersed-signal-data-2.sh 2 noshift
 
-newline
-  color 1 0 0
-  linethickness 2.0
-  linetype solid
+copycurve
   pts
     shell : ./dispersed-signal-data-2.sh 3 noshift
 

doc/papers/2011/europar/lofar.tex

@@ -229,7 +229,7 @@ Another major component in the pulsar-observation pipeline is real-time dedisper
 \end{minipage}
 \end{figure}
 
-Dedispersion is performed in the frequency domain, effectively by doing a 4K~Fourier transform (FFT) that splits a 12~KHz channel into 3~Hz subchannels. The phases of the observed samples are corrected by applying a Chirp function~\cite{...}, i.e., by multiplication with precomputed, subchannel-dependent, complex weights. These multiplications are programmed in assembly, to reduce the computational costs. A backward FFT is done to revert to 12~KHz channels. 
+Dedispersion is performed in the frequency domain, effectively by doing a 4K~Fourier transform (FFT) that splits a 12~KHz channel into 3~Hz subchannels. The phases of the observed samples are corrected by applying a chirp function~\cite{...}, i.e., by multiplication with precomputed, channel-dependent, complex weights. These multiplications are programmed in assembly, to reduce the computational costs. A backward FFT is done to revert to 12~KHz channels. 
 
 Figure~\ref{fig:dedispersion-result} shows the effectiveness of channel-level dedispersion, where we observed pulsar J0034-0534 with a pulse period of 1.88~ms. By applying dedispersion, the effective time resolution is improved from 0.51~ms to 0.082~ms, revealing a more detailed pulse and a better signal-to-noise ratio. Dedispersion thus contributes significantly to the data quality, but it also comes at a significant computational cost due to the two FFTs it requires. It demonstrates the power of using a \emph{software\/} telescope: the pipeline component was implemented, verified, and optimized in only one month time.