towards mobile phone localization without war-driving

Towards Mobile Phone Localization without War-Driving

Ionut Constandache, Romit Roy Choudhury, Injong Rhee

Location is an IP addressLocation is an IP addressfor content delivery

Location-Based Applications Location-Based Applications (LBAs)(LBAs)

Examples:

Location-based recommendations, geo-tagging

GeoLife: shopping list when near a grocery store

TrafficSense: real-time traffic conditions

Location expresses context of user◦Facilitates content delivery

Is GPS the solution?

Pros: Good accuracy Cons: Poor battery lifetime

Is GSM the solution?

Pros: Long battery lifetime Cons: Poor accuracy

What about WiFi Localization?

E.g., SkyHook:

Basic Idea:1. Several trucks war-drive a place2. Create Radio map = <Location: WiFi IDs>3. Distribute map to phones4. Phone user goes to war-driven region, overhears

WiFi IDs5. Reverse Look Up IDs against radio map6. Obtains location

Basic Idea:1. Several trucks war-drive a place2. Create Radio map = <Location: WiFi IDs>3. Distribute map to phones4. Phone user goes to war-driven region, overhears

WiFi IDs5. Reverse Look Up IDs against radio map6. Obtains location

Is Skyhook the solution?

Middle GroundLower Accuracy than GPS, Longer

Battery lifetime Better Accuracy than GSM, Shorter

Battery lifetimebut …

Is Skyhook the solution?

Middle GroundLower Accuracy than GPS, Longer Battery

lifetime Better Accuracy than GSM, Shorter Battery

lifetimebut …

At the cost of: Degraded location accuracy: walking paths ~

60m Reliance on infrastructure (APs) War-driving ($$ + carbon footprint) “NYTimes: Skyhook fleet 500 trucks/drivers”

ContentsContentsCompAcc

Evaluation

Limitations and Future Work

Conclusion

ContentsContentsCompAcc

Evaluation

Limitations and Future Work

Conclusion

GoalsGoals No War-Driving

Cannot drive walking paths (campus, parks, …)

Expensive / Environment unfriendly

No reliance on WiFi infrastructure Rural regions / developing countries

Good accuracy (~GPS)

Improve energy-efficiency Better than Skyhook, GPS

CompAcc: Basic IdeaCompAcc: Basic Idea Direction(compass) +

Displacement(accelerometer) = User’s directional trailDirectional Trail

CompAcc: Basic IdeaCompAcc: Basic Idea Direction(compass) +

Displacement(accelerometer) = User’s directional trail

Compute path signatures ◦ Derived from a local electronic map (Google Maps)

Path Signature …

CompAcc: Basic IdeaCompAcc: Basic Idea Direction(compass) + Displacement(accelerometer) =

User’s directional trail

Compute path signatures ◦ Derived from a local electronic map (Google Maps)

Compare directional trail with path signatures◦ Best match provides the user location

Directional Trail

Path Signature …

Path Signature … Directional Trail

Correct location errors at turns

Path Signature … Directional Trail

Correct location errors at turns

Directional Trail

Path Signature …

ArchitectureArchitectureTile

DatabaseTile

Database

6. Current location(lat A, long B)

2. Report initial location(lat X, long Y)

Tile

4. Direction(Compass)

5. Displacement(Accelerometer)

1. Initial location GPS:(lat X, long Y)

CompAcc

Initial location Directional trail Current location

3. Obtain paths in the user vicinity

Sample Tile:Sample Tile:

Directional trail: Directional trail: displacementdisplacementAccelerometer based step countdisplacement = step_count *

step_size

Directional trail: Directional trail: displacementdisplacementAccelerometer based step countdisplacement = step_count *

step_size

Directional trail: directionDirectional trail: direction

Directional trail: directionDirectional trail: direction

Path SignaturePath SignatureExtract from Google MapsGeodesic formulas

Matching Directional Trail Matching Directional Trail with Path Signatureswith Path Signatures

Dissimilarity Metric:

ci = compass readingspi = path computed directionN = directional trail size

Directional Trail

Path Signature

Fallback Mechanism: A-Fallback Mechanism: A-GPSGPSWhat if the dissimilarity metric is

large?◦Trigger A-GPS

Fallback MechanismFallback Mechanism

Estimated Location

7th Street

5th Street

Main

St.

AA

EE

Fallback MechanismFallback Mechanism

Estimated Location

7th Street

5th Street

Main

St.

AA

EE

ContentsContentsCompAcc

Evaluation

Limitations and Future Work

Conclusion

ResultsResultsCompared 3 localization schemes

◦CompAcc◦Skyhook◦Wifi-War-Walk (We war-droved walking

paths in campus)Metrics Instantaneous Error = distance(estimated, real)

Average Localization Error (ALE) = Average Instantaneous Error

CompAcc Instantaneous CompAcc Instantaneous ErrorError

ResultsResults Average ALE

GPS: 10mCompAcc: 11m

WiFi-War-Walk: 30mSkyhook: 70m

Energy GPS: 10h

CompAcc: 23hWiFi-War-Walk:16h

Skyhook:16h

Directional trail: Directional trail: displacementdisplacementStep count/displacement

accuracy

Trail/Path SizeTrail/Path Size

ContentsContentsCompAcc

Evaluation

Limitations and Future Work

Conclusion

Limitations and Future Limitations and Future WorkWork

Map Generation Manually mark footpaths

User Position Estimated along the Path Apply particle filters to accommodate wide

roads

Multiplexing between Localization Methods Hand-off to Skyhook/GPS when driving Extend to vehicular movement

ContentsContentsCompAcc

Evaluation

Limitations and Future Work

Conclusion

CompAccCompAccToday’s localization technologies

limited Energy- Efficiency Coverage/Accuracy

Rely on simple localization mechanism Need: Compass, Accelerometer and Maps

Evaluation results: ALE: 11m Battery: 23h

CompAcc scales to any mapped part of the world

AdvantagesAdvantages No war-driving No reliance on WiFi infrastructure

Maps available ubiquitously

Improves battery lifetime GPS ~10h Skyhook ~16h Accelerometer ~ 39h Compass ~48h