An Experimental Platform for Quantified Crowd M. Grabowski, M. - - PowerPoint PPT Presentation

An Experimental Platform for Quantified Crowd

• M. Grabowski, M. Marschall, W. Sirko,
• M. Debski, M. Ziombski, P. Horban,
• S. Acedanski, M. Peczarski, D. Batorski,
• K. Iwanicki

University of Warsaw

WiMAN 2015, Las Vegas, NV, USA, August 6th, 2015

Quantified Self

= Self knowledge through numbers

Image source: http://www.health2news.com/files/2011/11/WellnessFX-300x220.png Image source: http://www.peelapom.com/ketzirah/wp- content/uploads/2013/04/bodyMonitor_collage- filtered-1024x800.jpg

You can't manage what you don't measure

Quantified Crowd

The enabling factors:

• Further device

miniaturization;

• Low-power wireless inter-

device communication. The core ideas:

• Collaborative sensing of

collocated people.

• Online and (partially) in-

network data analysis.

• Coordinated feedback for

groups of people.

adapted from image: http://www.clker.com/clipart-smaller-crowd-rdc.html

Quantified Crowd

Crowd management:

• Detecting patterns of

coordinated behavior in a crowd.

• Automaticly recognizing and

predicting problematic situations.

• Notifying people and the

autorities about danger.

Image source: http://en.wikipedia.org/wiki/File:Crowd_in_street.jpg

You can't manage what you don't measure

Quantified Crowd

Maximizing event experience:

• Detecting interpersonal

interactions.

• Recognizing

communities.

• Navigation through

crowds.

• Contact and content

suggesting.

Image source: https://pixabay.com/en/music-kiss-rock-heavy-metal-819152/

Problem Statement

Prior research has been conducted with:

• Smartphones, or
• Custom devices.

We are not aware of a common platform that would allow for innovating at various levels

• f quantified crowd.

Image source: http://electronicdesign.com/content/14978/59382_fig_01.jpg

Experimental Platform

Goal:

• experimentation in as many scenarios as possible.

Components:

• Hardware
• Software

Hardware

Design goals:

• On its own should enable experimentation in

many scenarios.

• It should be possible to extended with external

devices.

Hardware

Based on ARM Cortex-M0+.

Hardware

Local sensors:

• Accelerometer
• Magnetometer
• Gyroscope
• Microphone
• Barometer (altimeter)
• Light sensor
• Thermometer
• Hygrometer

Hardware

Crowd texture sensors:

• Low-power 868MHz radio
• Infra-red transceiver
• Bluetooth Low Energy chip

Hardware

I/O interfaces:

• 400x300 e-paper display
• Buttons
• LEDs
• Buzzer
• External flash memory
• SD card
• USB socket

Software

• Badge software – written in NesC.
• External software – mostly Java.

Evaluation

• Conducted on a small scale.
• We are in the process of manufacturing a

large batch of badges.

Evaluation

Power consumption

Component Standby [μA] Active [mA] MCU 1.7 6.1 RF transceiver 0.2 16.9 (RX), 34.2 (TX @ 12 dBm) Infrared transceiver 0.9 (RX), 72 (TX @ power level 3) BLE module 0.4 25 (RX), 36 (TX) E-compass 2 0.44 Gyroscope 1 5 Barometer 0.5 0.025 Light sensor 4 0.13 Microphone 1.8 External flash 100 20 SD card depends on a card (e.g., 100) E-paper display 8

Evaluation

Proximity detection with RF

• Dist. [m]

Radio TX power [dBm]

• 30.2
• 27.7
• 20.5
• 15.7
• 11.0
• 5.0
• 0.3

3.7 1 74.8 85.3 99.0 99.7 99.4 99.3 99.7 99.6 2 14.0 65.7 99.7 99.2 99.0 99.6 99.3 98.4 3 2.0 0.0 98.7 98.9 99.3 99.6 99.2 99.3 5 14.9 0.0 99.4 99.1 99.6 99.4 97.9 99.3 10 0.0 0.0 36.0 96.0 99.7 99.9 99.0 99.3 20 0.0 0.0 0.0 14.9 98.3 99.9 99.3 99.5 30 0.0 0.0 0.0 0.0 23.6 35.0 99.0 99.1 40 0.0 0.0 0.0 0.0 20.0 22.0 68.5 99.5 50 0.0 0.0 0.0 0.0 0.0 1.0 11.0 98.8 60 0.0 0.0 0.0 0.0 0.0 0.0 23.4 98.1 70 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Evaluation

Proximity detection with IR

1 2 3 1 100.0 99.0 93.0 88.0 2 74.0 67.0 99.0 88.0 3 0.0 55.0 33.0 51.0 4 0.0 0.0 50.0 53.0 5 0.0 0.0 0.0 30.0 6 0.0 0.0 0.0 0.0

• Dist. [m]

IR TX power [level]

Evaluation

Mutual orientation detection with IR

Evaluation

Real-world situation detection

Conclusion

• Our platform has the potential to serve on its
• wn in various experimental scenarios for

quantified crowd.

• It can be extended with off-the-shelf mobile

devices, which further broadens its applicability.

• We are in the process of manufacturing 100

pieces of the badges.

• We hope that the platform will be used by
• ther groups.

Thank You

Questions?

Supported by the (Polish) National Science Centre (NCN) within the SONATA programme under grant no. DEC-2012/05/D/ST6/03582.

• K. Iwanicki was additionally supported by a scholarship from the (Polish)

Ministry of Science and Higher Education for outstanding young scientists.