EXPERIMENTS WITH MODEL-DRIVEN DATA ACQUISITION FOR CROWDSENSING - - PowerPoint PPT Presentation

▶

Aug 15, 2022 101 likes •240 views

EXPERIMENTS WITH MODEL-DRIVEN DATA ACQUISITION FOR CROWDSENSING 7/27/2013 Phillip Dold Socially Relevant REU Program 2013 Air Pollution You are concerned about air pollution in your city Finding the causes of the pollution Traditional

SLIDE 1

EXPERIMENTS WITH MODEL-DRIVEN DATA ACQUISITION FOR CROWDSENSING

Phillip Dold

7/27/2013

Socially Relevant REU Program 2013

SLIDE 2

2

Air Pollution

 You are concerned about air pollution in your city  Finding the causes of the pollution

 Traditional Setup

 Fixed Sensors

 Crowdsensing Solution?

 You could start a crowdsensing campaign

 Recruit friends, family, and strangers  Collect particulates per million

SLIDE 3

3

Crowdsensing

 Volunteers collect data with smartphones  Variety of sensors

 Accelerometer  GPS  Light Sensor  Microphone  Cameras

SLIDE 4

4

Challenges of crowdsensing

 Energy consumption

 Sensors require energy  Communication is one of the biggest energy drains

 Monetary Costs

 Mobile data plans are not free nor “unlimited”

 Both of these could decrease participation

SLIDE 5

5

Traditional Crowdsensing

Server / Database

Collect data with

Users

Sends data to

SLIDE 6

6

Model-Driven Data Acquisition

Server Model Learns

Queries phones

Researchers

Queries

Queries phones Users

Learning Phase

have

Query Phase Update Phase

Queries

SLIDE 7

7

Experimenting with Models

 Implemented a simulator in Java that can be used

to experiment with models and implementations

 Experimental Variables:

 Degree of mobility  Density of network  Type of data  Length of learning phase

 Evaluation of Metrics  Length of learning  Accuracy of model  Number of Updates

SLIDE 8

8

DBP (Derivative Based Predictions) [Raza 2012]

Simple Time Series

Model

Simpler Calculations
Less data needed

Expectation: Performance will drop with mobility

SLIDE 9

9

DrOPS (model-Driven Optimizations for Public Sensing )[Philipp 2013]

Multivariate Gaussian

Distribution Model

More Complex Calculations
More data needed

Expectation: Model will perform well, but will consume more energy than DBP

SLIDE 10

10

Experimental Setup

 Simulator built in Java  Estimates Energy usage

 Communication  Sensors

 Datasets  Intel Lab  Lausanne Urban

Canopy Experiment

 Mobility Traces

 Cab spotting data from Crawdad

Intel sensor lab

SLIDE 11

11

Conclusions

 Model-Driven Data Acquisition  Building a model rather than constantly sending data  It can help reduce communication  The simulator is still under development  Looking for additional data sets to use

SLIDE 12

12

Questions?

SLIDE 13

13

References

Philipp, D., Stachowiak, J., Alt, P., Durr, F., and Rothermel, K. DrOPS: Model-Driven Optimization for Public Sensing Systems. In 2013 IEEE International Conference on Pervasive Computing and Communications (PerCom) (PerCom 2013) (San Diego, CA, USA, March 2013), IEEE Computer Society, pp. 1-8. Raza, U., Camerra, A., Murphy, A. L., Palpanas, T., and Picco, G. P. What does model-driven data acquisition really achieve in wireless sensor networks? In Pervasive Computing and Communications (PerCom), 2012 IEEE International Conference on (2012), IEEE, pp. 85-94.