Presentation is loading. Please wait.

Presentation is loading. Please wait.

Sensing Meets Mobile Social Networks: The Design, Implementation and Evaluation of the CenceMe Application Emiliano Miluzzo†, Nicholas D. Lane†, Kristóf.

Similar presentations


Presentation on theme: "Sensing Meets Mobile Social Networks: The Design, Implementation and Evaluation of the CenceMe Application Emiliano Miluzzo†, Nicholas D. Lane†, Kristóf."— Presentation transcript:

1 Sensing Meets Mobile Social Networks: The Design, Implementation and Evaluation of the CenceMe Application Emiliano Miluzzo†, Nicholas D. Lane†, Kristóf Fodor†, Ronald Peterson†, Hong Lu†, Mirco Musolesi†, Shane B. Eisenman§, Xiao Zheng†, Andrew T. Campbell† †Computer Science, Dartmouth College §Electrical Engineering, Columbia University Slides from http://nslab.ee.ntu.edu.tw/NetworkSeminar/slides/cenceme.ppt

2 Outline Introduction of CenceMe Design Consideration CenceMe Implementation CenceMe Classifier System Performance User Study Conclusion

3 Motivation Text messaging: “Where are you?” “What are you doing?” Sensors in mobile phone: GPS, accelerometers, microphone, camera … etc Data collection through sensors

4 Introduction of CenceMe People-centric sensing application Implementation on Nokia N95; Symbian/JME VM platform Share user presence information (Facebook)

5 Contributions Design, implementation and evaluation Lightweight classifier Trade-off: time fidelity v.s. latency Complete User study

6 Mobile Phone Limitations OS Limitations API and Operational Limitations Security Limitations Energy Management Limitations

7 CenceMe Architecture

8 Architecture Design Issues Split-Level Classification (primitives, facts) – Customized tag – Resiliency – Minimize bandwidth usage/energy – Privacy/data integrity Power Aware Duty-Cycle

9 CenceMe Implementation Operations (Phone): Sensing Classification to produce primitives Presentation of people's presence on the phone Upload of primitives to backend servers Classifications (Backend Server): Classifying the nature of the sound collected from the microphone Classifying the accelerometer data to determine activity (sitting, standing, walking, running) Scanned Bluetooth/MAC addresses in range GPS readings Random photos

10 Phone Software

11 ClickStatus

12 Backend Software

13 Phone classifiers (1/2) Audio – Feature extraction – Classification Human voiceEnvironment noise

14 Mean Standard Deviation

15 Phone classifiers (2/2) Activity Sitting Standing Walking Running Time

16 Backend Classifier Conversation Social context – Neighborhood conditions – Social status Mobility mode detector (vehicle or not) Location (to description/icon) “Am I Hot” – Nerdy, party animal, cultured, healthy, greeny

17 System Performance Classifier accuracy Impact of mobile phone placement on body – 8 users – Annotations as ground truth for comparison with classifier outputs Environmental conditions Sensing duty cycles

18 General Results

19 Phone Placement on Body Pocket, lanyard, clipped to belt Insignificant impact conversation vs. Non-conversation

20 Environmental Impacts Independent of activity classification More important: transition between locations

21 Duty Cycle Problem detecting short term event Experiment: 8 people. Reprogram different duty cycles

22 Power Benchmarks Measuring battery voltage, current, temperature Battery lifetime: 6.22+/- 0.59 hours

23 Memory and CPU Benchmarks

24 User Study Survey user experience Feedback: – Positive from all users – Willing to share detail status and presence information on Facebook – Privacy not an issue (??) – Stimulate curiosity among users – Self-learning on activity patterns and social status

25 User Study A new way to connect people What is the potential CenceMe demographic? Learn about yourself and your friends My friends always with me

26 Conclusion A complete design, implementation and evaluation First application to retrieve and publish sensing presence A complete user study and feedback for future improvement


Download ppt "Sensing Meets Mobile Social Networks: The Design, Implementation and Evaluation of the CenceMe Application Emiliano Miluzzo†, Nicholas D. Lane†, Kristóf."

Similar presentations


Ads by Google