jfriedt / IFCS2018 article

Download as
Browse Code ยป

Commit 970e2bac624cbac378f28f6424e5bbff0e5750ad

Authored by ahugeat
1 parent 315be2a309
Exists in master

Ajout des valeurs de rejections dans le tableau.

Showing 2 changed files with 23 additions and 16 deletions Side-by-side Diff

  1 +ifcs2018.aux
  2 +ifcs2018.bbl
  3 +ifcs2018.blg
  4 +ifcs2018.log
  5 +ifcs2018.out
  6 +ifcs2018.pdf
  7 +*.bak
... ... @@ -13,8 +13,8 @@
13 13 \title{Filter optimization for real time digital processing of radiofrequency signals: application
14 14 to oscillator metrology}
15 15  
16   -\author{\IEEEauthorblockN{A. Hugeat\IEEEauthorrefmark{1}\IEEEauthorrefmark{2}, J. Bernard\IEEEauthorrefmark{2},
17   -G. Goavec-M\'erou\IEEEauthorrefmark{1},
  16 +\author{\IEEEauthorblockN{A. Hugeat\IEEEauthorrefmark{1}\IEEEauthorrefmark{2}, J. Bernard\IEEEauthorrefmark{2},
  17 +G. Goavec-M\'erou\IEEEauthorrefmark{1},
18 18 P.-Y. Bourgeois\IEEEauthorrefmark{1}, J.-M Friedt\IEEEauthorrefmark{1}}
19 19 \IEEEauthorblockA{\IEEEauthorrefmark{1}FEMTO-ST, Time \& Frequency department, Besan\c con, France }
20 20 \IEEEauthorblockA{\IEEEauthorrefmark{2}FEMTO-ST, Computer Science department DISC, Besan\c con, France \\
21 21  
... ... @@ -41,10 +41,10 @@
41 41 \section{Digital signal processing of ultrastable clock signals}
42 42  
43 43 Analog oscillator phase noise characteristics are classically performed by downconverting
44   -the radiofrequency signal using a saturated mixer to bring the radiofrequency signal to baseband,
  44 +the radiofrequency signal using a saturated mixer to bring the radiofrequency signal to baseband,
45 45 followed by a Fourier analysis of the beat signal to analyze phase fluctuations close to carrier. In
46 46 a fully digital approach, the radiofrequency signal is digitized and numerically downconverted by
47   -multiplying the samples with a local numerically controlled oscillator (Fig. \ref{schema}) \cite{rsi}.
  47 +multiplying the samples with a local numerically controlled oscillator (Fig. \ref{schema}) \cite{rsi}.
48 48  
49 49 \begin{figure}[h!tb]
50 50 \begin{center}
... ... @@ -89,10 +89,10 @@
89 89 resources \cite{yu2007design, kodek1980design}.
90 90  
91 91 The degrees of freedom when addressing the problem of replacing the single monolithic
92   -FIR with a cascade of optimized filters are the number of coefficients $N_i$ of each filter $i$,
93   -the number of bits $c_i$ representing the coefficients and the number of bits $d_i$ representing
94   -the data fed to the filter. Because each FIR in the chain is fed the output of the previous stage,
95   -the optimization of the complete processing chain within a constrained resource environment is not
  92 +FIR with a cascade of optimized filters are the number of coefficients $N_i$ of each filter $i$,
  93 +the number of bits $c_i$ representing the coefficients and the number of bits $d_i$ representing
  94 +the data fed to the filter. Because each FIR in the chain is fed the output of the previous stage,
  95 +the optimization of the complete processing chain within a constrained resource environment is not
96 96 trivial. The resource occupation of a FIR filter is considered as $c_i+d_i+\log_2(N_i)$ which is
97 97 the number of bits needed in a worst case condition to represent the output of the FIR.
98 98  
... ... @@ -118,7 +118,7 @@
118 118  
119 119 The MILP solver provides a solution to the problem by selecting a series of small FIR with
120 120 increasing number of bits representing data and coefficients as well as an increasing number
121   -of coefficients, instead of a single monolithic filter. Fig. \ref{compare-fir} exhibits the
  121 +of coefficients, instead of a single monolithic filter. Fig. \ref{compare-fir} exhibits the
122 122 performance comparison between one solution and a monolithic FIR when selecting a cutoff
123 123 frequency of half the Nyquist frequency: a series of 5 FIR and a series of 10 FIR with the
124 124 same space usage are provided as selected by the MILP solver. The FIR cascade provides improved
... ... @@ -143,9 +143,9 @@
143 143 \begin{center}
144 144 \begin{tabular}{|c|cccc|}\hline
145 145 FIR & BlockRAM & LookUpTables & DSP & rejection (dB)\\\hline\hline
146   -1 (monolithic) & 1 & 4064 & 40 & \\
147   -5 & 5 & 12332 & 0 & \\
148   -10 & 10 & 12717 & 0 & \\\hline\hline
  146 +1 (monolithic) & 1 & 4064 & 40 & -71.78 \\
  147 +5 & 5 & 12332 & 0 & -216.58 \\
  148 +10 & 10 & 12717 & 0 & -251.01 \\\hline\hline
149 149 Zynq 7010 & 60 & 17600 & 80 & \\\hline
150 150 \end{tabular}
151 151 \end{center}
152 152  
... ... @@ -181,14 +181,14 @@
181 181 characteristics of stable oscillators. The flexibility of the digital approach makes the result
182 182 best suited for closing the loop and using the measurement output in a feedback loop for
183 183 controlling clocks, e.g. in a quartz-stabilized high performance clock whose long term behavior
184   -is controlled by non-piezoelectric resonator (sapphire resonator, microwave or optical
  184 +is controlled by non-piezoelectric resonator (sapphire resonator, microwave or optical
185 185 atomic transition).
186 186  
187 187 \section*{Acknowledgement}
188 188  
189   -This work is supported by the ANR Programme d'Investissement d'Avenir in
190   -progress at the Time and Frequency Departments of the FEMTO-ST Institute
191   -(Oscillator IMP, First-TF and Refimeve+), and by R\'egion de Franche-Comt\'e.
  189 +This work is supported by the ANR Programme d'Investissement d'Avenir in
  190 +progress at the Time and Frequency Departments of the FEMTO-ST Institute
  191 +(Oscillator IMP, First-TF and Refimeve+), and by R\'egion de Franche-Comt\'e.
192 192 The authors would like to thank E. Rubiola, F. Vernotte, G. Cabodevila for support and
193 193 fruitful discussions.
194 194