CH3Br, 201009 (2+n)REMPI spectra assignments for region 66000 – 70000 cm-1
r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1,...
Transcript of r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1,...
![Page 1: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/1.jpg)
earthquake early warning using smartphones
A collaboration between UC Berkeley Seismological Laboratory,
Deutsche Telekom’s Silicon Valley Innovation Lab, and Sense Observation Systems (Sense OS)
Dr. Richard Allen, Qingkai Kong, Steven Allen, Dr. Jennifer Strauss
UC Berkeley Seismological Laboratory Dr. Young-Woo Kwon, Utah State University
Louis Schreier, Deutsche Telekom Silicon Valley Innovation Lab Ted Schmidt, Sense OS
![Page 2: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/2.jpg)
earthquake early warning (eew) • Global and unpredictable • Seismic networks only beginning to
provide early warning • Limit to the number of sensors • Greatest risks: populated areas,
industrial facilities, transportation systems, hospitals and schools
Innovative Science • Turn commodity smartphones
into sensors & warning devices • Smartphone penetration
matches population density – Greatest coverage in areas of
greatest risk
• Model for future sensors and early warning networks
Picture courtesy: John Elliott
![Page 3: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/3.jpg)
a simple idea
BUT, Smartphones are black boxes: • Consumer technology • Various manufacturers • Different applications, state • Connected or battery powered
3
Classified by the neural network algorithm developed by UC Berkeley
Captured by smartphone’s sensors Activity-based data
And generate an Early Warning or Not
![Page 4: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/4.jpg)
is it feasible?
![Page 5: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/5.jpg)
characterize the noise floor
![Page 6: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/6.jpg)
compare reference accelerometer
![Page 7: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/7.jpg)
capture human motion and shake data
![Page 8: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/8.jpg)
CcHhEeCcKkĸ TtHhEe CcLlOoCcKkĸ’Ss sampling accuracy
GS4A-R1 GS4A-R2 GS4B-R1 GS4B-R2GS4A-R1 1GS4A-R2 0.978 1GS4B-R1 0.998 0.962 1GS4B-R2 0.987 0.932 0.995 1
(msec) first second third fourth fifth sixthmin 0 0 0 1 0 0max 51 49 47 52 51 41
median 20 20 20 20 20 20average 20.20 20.19 20.19 20.19 20.19 20.19variance 0.48 0.43 0.41 0.45 0.42 0.46
std dev 0.69 0.65 0.64 0.67 0.65 0.68
![Page 9: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/9.jpg)
WwHhIiıCcHh BbRrIiıNnGgSs UuSs TtOo ….
9
classification accuracy
Maximum number of zero crossings
Pe
ak a
cce
lera
tio
n
(ve
cto
r su
m)
Earthquakes
Initial Results: 99.8% earthquake
Windows were classified as earthquakes
Feat
ure
1
Feature 2
![Page 10: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/10.jpg)
and, an opportunistic event
Above: Berkeley, CA • 38 km away – the Napa
‘Quake was recorded on a MyShake Smartphone
Napa, CA. • August 2014, 3 AM PST:
A 6.0 Magnitude Earthquake Struck South Napa.
A nearby Seismic Station’s recording
![Page 11: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/11.jpg)
http://seismo.berkeley.edu/~myShake/AppServer/
Smartphone seismic network
ongoing this semester: beta trials at berkeley
• scale users • data collection & analysis • app and system behavior
![Page 12: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/12.jpg)
Picture courtesy: John Elliott
evolution
AaNnDd …
![Page 13: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/13.jpg)
a vision of future eew networks
connected to a high-performance, intelligent, back-end providing real-time triggering, classification and alerts
tens of millions of smartphones, very low-cost, high capability processors, and smart internet enabled appliances
![Page 14: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/14.jpg)
Thank You [email protected], And if you want to sign up (Android only, please), email:
![Page 15: r ry g- ty ab S - Stanford Universityweb.stanford.edu/group/scpnt/pnt/PNT14/2014... · MotoX1, 90000 80000 70000 60000 50000 40000 30000 20000 10000 00 , 100k Native, Battery Samples](https://reader030.fdocuments.us/reader030/viewer/2022041013/5ec3e7cff487bc72e73184b4/html5/thumbnails/15.jpg)
a quick background
• UC Berkeley Seismology Laboratory is part of the California Integrated Seismic Networks (CISN)
• UCB’s Earthquake Early Warning research is sponsored by – Moore Foundation, – the USGS, – google among others
• Deutsche Telekom, Silicon Valley contributes funding, smartphones, and smartphone s/w engineering to UCB’s Seismology Lab
• MyShake: A UC Berkeley / DT collaboration in advanced research to integrate traditional seismic, commodity and mobile technologies for the purpose of developing a large, dynamic EEW network based on smartphones