PowerPoint Presentation - Introduction to Bicoherence Monitors
Download
Report
Transcript PowerPoint Presentation - Introduction to Bicoherence Monitors
HW
LIG
S
O
Introduction to
Bicoherence Monitors:
BicoViewer & BicoMon
Steve Penn (HWS)
HW
LIG
S
O
Why use BicoViewer?
Noise Investigations
» Step 1: Power Spectral Density (PSD) investigation, done in DTT
– Tells you the frequency and power of the noise
– If the noise is a “known” source, then you are done
» Step 2: Coherence investigation , done in DTT
– Tells you the phase correlation between two channels
– This gives an indication of the physical coupling between these channel
source
» Step 3: Bicoherence investigation, done in BicoViewer
– Tells you the phase correlation phase correlation at different frequencies,
possibly different channels
– Indicates the coupling either in the physical system or in the sensing
LSC Meeting • August 2004
LIGO DCC G040410-00-Z
2
HW
LIG
S
O
Statistics (1)
“1D” Statistics: (2nd Order Cumulants, 1st Order Spectra)
*
» Correlation: Cxy t x yt d X f Y f Sxy f
» Power Spectral Density: C2 x t
» Coherence:
X f X * f S2x f
Sxy f
C xy f
S2 x f S2y f
– Tells us power and phase coherence at a given frequency
LSC Meeting • August 2004
LIGO DCC G040410-00-Z
3
HW
LIG
S
O
Statistics (2)
“2D” Statistics: (3rd Order Cumulants, 2nd Order Spectra)
» Bicumulant:
Cxyz t, t x yt z t d X f1 Y f2 Z * f1 f2 Sxyz f1 , f2
*
» Bispectral Density: C3x t X f1 X f2 X f1 f2 S3x f1 , f2
» Bicoherence:
Cxyz f
Cxyz f
Sxyz f1, f 2
Sxx f1 Syy f 2 Szz f1 f 2
Sxyz f1, f 2
Sxx f1 Syy f 2 Szz f1 f 2
– Tells us power and phase coherence at a coupled frequency
LSC Meeting • August 2004
LIGO DCC G040410-00-Z
4
HW
LIG
S
O
Why Higher Order Statistics?
For a Gaussian process:Cnx t 0,
For independent processes:
z t x t yt ,
for n 2
Cnz t Cnx t Cny t Cnyt
n2
Allows for isolation of nonGaussian processes
» Visual check of frequency coupling and phase noise
» Statistical test for the probability of gaussianity and linearity
» Iterative process to reconstruct nongaussian signal from the higher
order cumulants
LSC Meeting • August 2004
LIGO DCC G040410-00-Z
5
HW
LIG
S
O
BicoViewer (Foreground Monitor)
Plots Bicoherence (auto- or cross-) & PSD’s of input channels
Proper decimation, Optimized windowing, Vectorized FFT
User selects: (Channels, Frequency Range, Frequency resolution, Data Span, Overlap)
Outputs GIF files. Script to make video.
Table of Current and New Parameters with Accuracy & Time Estimates
Plot Countdown clock
Text File Output
Faster Smoothing Routine
Heterodyning
Strip Chart for Monitoring Bicoherence of certain ROI
Output calculation parameters as configuration for BicoMon
LSC Meeting • August 2004
LIGO DCC G040410-00-Z
6
Current Version
LSC Meeting • August 2004
LIGO DCC G040410-00-Z
HW
LIG
S
O
7
Current Version (2)
LSC Meeting • August 2004
LIGO DCC G040410-00-Z
HW
LIG
S
O
8
HW
LIG
S
O
H1:LSC-AS_I movie
QuickTime™ and a
GIF decompressor
are needed to see this picture.
LSC Meeting • August 2004
LIGO DCC G040410-00-Z
9
HW
LIG
S
O
H1:LSC-AS_Q movie
QuickTime™ and a
GIF decompressor
are needed to see this picture.
LSC Meeting • August 2004
LIGO DCC G040410-00-Z
10
HW
LIG
S
O
Online Demo
(requires X windows)
On Fortress:
> /export/home/spenn/BicoViewer
new version soon in DMT
LSC Meeting • August 2004
LIGO DCC G040410-00-Z
11
HW
LIG
S
O
BicoMon: Current Version
• Monitor integrates bicoherence over specified frequency ROI
• Calculates bicoherence for multiple channel combinations
• Integrates bicoherence over multiple specified ROI
• Integrates bicoherence over entire unique area (Gaussianity)
• Trends Data and sends to DMTviewer.
• Included proper decimation with filtering rather than rebinning
• Tests for Operational State
LSC Meeting • August 2004
LIGO DCC G040410-00-Z
12
3 30
C 2 H1:LSC-AS_Q
H1:SUS-ITMX_OPLEV_PERROR 2048
256
1.0
0.5
Configuration File
Calculation
Parameters
Measurement
Parameters
M
M
M
M
M
M
M
M
M
M
M
M
M
LSC Meeting • August 2004
16384
64
Bico:H1:AS_Q-ITMY_OPLEV_PERROR_ALL
0 00
Bico:H1:AS_Q-ITMY_OPLEV_PERROR_60_2_2 60 2 2
Bico:H1:AS_Q-ITMY_OPLEV_PERROR_60_10_2 60 10 2
Bico:H1:AS_Q-ITMY_OPLEV_PERROR_60_38_2 60 38 2
Bico:H1:AS_Q-ITMY_OPLEV_PERROR_60_50_2 60 50 2
Bico:H1:AS_Q-ITMY_OPLEV_PERROR_120_2_2 120 2 2
Bico:H1:AS_Q-ITMY_OPLEV_PERROR_120_10_2 120 10 2
Bico:H1:AS_Q-ITMY_OPLEV_PERROR_120_38_2 120 38 2
Bico:H1:AS_Q-ITMY_OPLEV_PERROR_120_50_2 120 50 2
Bico:H1:AS_Q-ITMY_OPLEV_PERROR_180_2_2 180 2 2
Bico:H1:AS_Q-ITMY_OPLEV_PERROR_180_10_2 180 10 2
Bico:H1:AS_Q-ITMY_OPLEV_PERROR_180_38_2 180 38 2
Bico:H1:AS_Q-ITMY_OPLEV_PERROR_180_50_2 180 50 2
C 1 H1:LSC-AS_Q
256
1.0
M
M
M
M
O
64
Bico:H1:AS_Q-ITMX_OPLEV_PERROR_ALL
0 00
Bico:H1:AS_Q-ITMX_OPLEV_PERROR_60_2_2 60 2 2
Bico:H1:AS_Q-ITMX_OPLEV_PERROR_60_10_2 60 10 2
Bico:H1:AS_Q-ITMX_OPLEV_PERROR_60_38_2 60 38 2
Bico:H1:AS_Q-ITMX_OPLEV_PERROR_60_50_2 60 50 2
Bico:H1:AS_Q-ITMX_OPLEV_PERROR_120_2_2 120 2 2
Bico:H1:AS_Q-ITMX_OPLEV_PERROR_120_10_2 120 10 2
Bico:H1:AS_Q-ITMX_OPLEV_PERROR_120_38_2 120 38 2
Bico:H1:AS_Q-ITMX_OPLEV_PERROR_120_50_2 120 50 2
Bico:H1:AS_Q-ITMX_OPLEV_PERROR_180_2_2 180 2 2
Bico:H1:AS_Q-ITMX_OPLEV_PERROR_180_10_2 180 10 2
Bico:H1:AS_Q-ITMX_OPLEV_PERROR_180_38_2 180 38 2
Bico:H1:AS_Q-ITMX_OPLEV_PERROR_180_50_2 180 50 2
C 2 H1:LSC-AS_Q
H1:SUS-ITMY_OPLEV_PERROR 2048
256
1.0
0.5
M
M
M
M
M
M
M
M
M
M
M
M
M
16384
HW
LIG
S
0.5
16384
64
Bico:H1:AS_Q_ALL
0 00
Bico:H1:AS_Q_120_2_2 120 2 2
Bico:H1:AS_Q_120_10_2 120 10 2
Bico:H1:AS_Q_120_38_2 120 38 2
LIGO DCC G040410-00-Z
13
HW
LIG
S
O
Conclusions
Bicoherence monitors could be a useful tool for analyzing data
for glitches, gaussianity, upconversion, and chirps.
We are now at sensitivity where up-converted data can be seen
BicoMon operational.
BicoViewer works (Beta Testers Needed).
Beta Test Need: Now that the IFO is at/near a sensitivity to
see bicoherence, we need people to use monitor and make
bicoherence an analysis tool on par with PSDs or coherence.
LSC Meeting • August 2004
LIGO DCC G040410-00-Z
14