Download - Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:
![Page 1: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/1.jpg)
1
Energy-efficient Localization Via Personal Mobility Profiling
Ionut Constandache
Co-authors: Shravan Gaonkar, Matt Sayler, Romit Roy Choudhury and Landon
Cox
![Page 2: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/2.jpg)
2
Context
Pervasive wireless connectivity+
Localization technology=
Location-based applications (LBAs) Location-based applications (LBAs)
![Page 3: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/3.jpg)
3
Context
Pervasive wireless connectivity+
Localization technology=
(iPhone AppStore: 3000 LBAs, Android: 600
LBAs)
Location-based applications (LBAs) Location-based applications (LBAs)
![Page 4: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/4.jpg)
4
Location-Based Applications (LBAs)
Two kinds of LBAs: One-time location information:Geo-tagging, location-based recommendations, etc.
![Page 5: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/5.jpg)
5
Location-Based Applications (LBAs)
Two kinds of LBAs: One-time location information:Geo-tagging, location-based recommendations, etc.
Localization over long periods of time:GeoLife: shopping list when near a grocery storeTrafficSense: real-time traffic conditions
![Page 6: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/6.jpg)
6
Localization Technology
LBAs rely on localization technology to get user position
![Page 7: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/7.jpg)
7
Localization Technology
LBAs rely on localization technology to get user position
Accuracy Technology
10m GPS 20-40m
WiFi 200-400m GSM
Accuracy Technology
10m GPS 20-40m
WiFi 200-400m GSM
![Page 8: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/8.jpg)
8
Localization Technology
LBAs rely on localization technology to get user position
LBAs executed on mobile phones
Accuracy Technology
10m GPS 20-40m
WiFi 200-400m GSM
Accuracy Technology
10m GPS 20-40m
WiFi 200-400m GSM
![Page 9: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/9.jpg)
9
Localization Technology
LBAs rely on localization technology to get user position
LBAs executed on mobile phones
Accuracy Technology
10m GPS 20-40m
WiFi 200-400m GSM
Accuracy Technology
10m GPS 20-40m
WiFi 200-400m GSM
Energy Efficiency is importantEnergy Efficiency is important (localization for long time)
![Page 10: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/10.jpg)
10
Localization Technology
Ideally Accurate and Energy-Efficient Localization
Ideally Accurate and Energy-Efficient Localization
![Page 11: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/11.jpg)
11
Energy
… sample every 30s
Battery Lifetime: GPS ~ 10h WiFi ~ 40h GSM ~ 60h
Battery Lifetime: GPS ~ 10h WiFi ~ 40h GSM ~ 60h
![Page 12: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/12.jpg)
12
Energy
… sample every 30s
Battery shared with Talk time, web browsing, photos, SMS, etc.
Battery Lifetime: GPS ~ 10h WiFi ~ 40h GSM ~ 60h
Battery Lifetime: GPS ~ 10h WiFi ~ 40h GSM ~ 60h
![Page 13: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/13.jpg)
13
Energy
… sample every 30s
Battery shared with Talk time, web browsing, photos, SMS, etc.
Localization energy budget only percentage of battery 20% of battery = 2h GPS or 8h WiFi
Battery Lifetime: GPS ~ 10h WiFi ~ 40h GSM ~ 60h
Battery Lifetime: GPS ~ 10h WiFi ~ 40h GSM ~ 60h
![Page 14: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/14.jpg)
14
Energy
… sample every 30s
Battery shared with Talk time, web browsing, photos, SMS, etc.
Localization energy budget only percentage of battery 20% of battery = 2h GPS or 8h WiFi
Battery Lifetime: GPS ~ 10h WiFi ~ 40h GSM ~ 60h
Battery Lifetime: GPS ~ 10h WiFi ~ 40h GSM ~ 60h
For limited energy budget what accuracy to expect?For limited energy budget what accuracy to expect?
![Page 15: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/15.jpg)
15
L(t0) L(t1) L(t2) L(t3) L(t4)
L(t6)L(t7)
L(t5)
Problem Formulation
![Page 16: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/16.jpg)
16
L(t0) L(t1) L(t2) L(t3) L(t4)
L(t6)L(t7)
L(t5)
LocalizationError
t0 t1 t2 t3 t4 t5 t6 t7 Time
Problem Formulation
![Page 17: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/17.jpg)
17
L(t0) L(t1) L(t2) L(t3) L(t4)
L(t6)L(t7)
L(t5)
LocalizationError
t0 t1 t2 t3 t4 t5 t6 t7 Time
GPS
Problem Formulation
![Page 18: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/18.jpg)
18
L(t0) L(t1) L(t2) L(t3) L(t4)
L(t6)L(t7)
L(t5)
LocalizationError
t0 t1 t2 t3 t4 t5 t6 t7 Time
GPS
Problem Formulation
Accuracy gain from GPSEng.: 1 GPS read
![Page 19: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/19.jpg)
19
L(t0) L(t1) L(t2) L(t3) L(t4)
L(t6)L(t7)
L(t5)
LocalizationError
t0 t1 t2 t3 t4 t5 t6 t7 Time
GPS
Accuracy gain from GPSEng.: 1 GPS read
Problem Formulation
Accuracy gain from WiFiEng.: 1 WiFi read
WiFi
![Page 20: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/20.jpg)
20
L(t0) L(t1) L(t2) L(t3) L(t4)
L(t6)L(t7)
L(t5)
LocalizationError
t0 t1 t2 t3 t4 t5 t6 t7 Time
GPS
Accuracy gain from GPSEng.: 1 GPS read
Problem Formulation
Accuracy gain from WiFiEng.: 1 WiFi read
WiFi
![Page 21: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/21.jpg)
21
Given energy budget B, known Trace T, location readings costs egps , ewifi , egsm :
Schedule location readings to minimize Average Localization Error (ALE)
Given energy budget B, known Trace T, location readings costs egps , ewifi , egsm :
Schedule location readings to minimize Average Localization Error (ALE)
Problem Formulation
![Page 22: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/22.jpg)
22
Given energy budget B, known Trace T, location readings costs egps , ewifi , egsm :
Schedule location readings to minimize Average Localization Error (ALE)
Given energy budget B, known Trace T, location readings costs egps , ewifi , egsm :
Schedule location readings to minimize Average Localization Error (ALE)
Problem Formulation
ALE = Avg. dist. between reported and actual location of the user
![Page 23: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/23.jpg)
23
Given energy budget B, known Trace T, location readings costs egps , ewifi , egsm :
Schedule location readings to minimize Average Localization Error (ALE)
Given energy budget B, known Trace T, location readings costs egps , ewifi , egsm :
Schedule location readings to minimize Average Localization Error (ALE)
Problem Formulation
ALE = Avg. dist. between reported and actual location of the user
Find the Offline Optimal AccuracyFind the Offline Optimal Accuracy
![Page 24: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/24.jpg)
24
Results
B = 25% BatteryOpt. GPS/WiFi/GSM
Trace 1 78.5m
Trace 2 58.6m
Trace 3 62.1m
![Page 25: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/25.jpg)
25
B = 25% BatteryOpt. GPS/WiFi/GSM
Trace 1 78.5m
Trace 2 58.6m
Trace 3 62.1m
Offline Optimal ALE > 60mOffline Optimal ALE > 60m
Results
![Page 26: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/26.jpg)
26
Offline Optimal ALE > 60mOffline Optimal ALE > 60m
Results
Online Schemes Naturally WorseOnline Schemes Naturally Worse
B = 25% BatteryOpt. GPS/WiFi/GSM
Trace 1 78.5m
Trace 2 58.6m
Trace 3 62.1m
![Page 27: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/27.jpg)
27
Our Approach: EnLoc
Reporting last sampled location increases inaccuracy
![Page 28: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/28.jpg)
28
Our Approach: EnLoc
Reporting last sampled location increases inaccuracy
Prediction opportunities exist Exploit habitual paths Leverage population statistics when the user has deviated
![Page 29: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/29.jpg)
29
Our Approach: EnLoc
Reporting last sampled location increases inaccuracy
Prediction opportunities exist Exploit habitual paths Leverage population statistics when the user has deviated
EnLoc Solution: Predict user location when not sampling Sample when prediction is unreliable
![Page 30: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/30.jpg)
30
EnLoc: Overview
Deviations
EnLoc
Habitual Paths
E.g. Regular path to office E.g. Going to a vacation
![Page 31: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/31.jpg)
31
EnLoc: Overview
Deviations
EnLoc
Habitual Paths
E.g. Regular path to office
Per-user Mobility ProfilePer-user Mobility Profile
E.g. Going to a vacation
![Page 32: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/32.jpg)
32
EnLoc: Overview
Deviations
EnLoc
Habitual Paths
E.g. Regular path to office E.g. Going to a vacation
Per-user Mobility ProfilePer-user Mobility Profile Population StatisticsPopulation Statistics
![Page 33: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/33.jpg)
33
Profiling Habitual Mobility
Intuition: Humans have habitual activities Going to/from office Favorite grocery shop, cafeteria
![Page 34: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/34.jpg)
34
Profiling Habitual Mobility
Intuition: Humans have habitual activities Going to/from office Favorite grocery shop, cafeteria
Habitual activities translate into habitual paths E.g. path from home to office
![Page 35: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/35.jpg)
35
Profiling Habitual Mobility
Intuition: Humans have habitual activities Going to/from office Favorite grocery shop, cafeteria
Habitual activities translate into habitual paths E.g. path from home to office
Habitual paths may branch E.g., left for office, right for grocery Q: How to solve uncertainty? A: Schedule a location reading after the branching point.
![Page 36: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/36.jpg)
36
Per-User Mobility Graph
User Habitual Paths
Graph of habitual visited GPS coordinates
![Page 37: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/37.jpg)
37
Per-User Mobility Graph
User Habitual Paths Logical Representation
Graph of habitual visited GPS coordinates
![Page 38: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/38.jpg)
38
Per-User Mobility Graph
Graph of habitual visited GPS coordinates Sample location after branching points Predict between branching points # of BPs < # of location samples(BP = branching point)
User Habitual Paths Logical Representation
![Page 39: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/39.jpg)
39
Evaluation: Habitual Paths
30 days of traces, loc. battery budget 25% per day
Assume phone speed known
![Page 40: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/40.jpg)
40
Evaluation: Habitual Paths
30 days of traces, loc. battery budget 25% per day
Assume phone speed known
![Page 41: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/41.jpg)
41
Evaluation: Habitual Paths
30 days of traces, loc. battery budget 25% per day
Assume phone speed known
Average ALE 12m
![Page 42: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/42.jpg)
42
Predict based on population statistics If user on a certain street, at the next intersection
predict the most probable turn.
Deviations from habitual paths
![Page 43: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/43.jpg)
43
Predict based on population statistics If user on a certain street, at the next intersection
predict the most probable turn. Probability Maps computed from Google Map simulation
Deviations from habitual paths
![Page 44: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/44.jpg)
44
Predict based on population statistics If user on a certain street, at the next intersection
predict the most probable turn. Probability Maps computed from Google Map simulation
Deviations from habitual paths
Goodwin & Green
U-Turn Straight Right Left
E on Green 0 0.881 0.039 0.078
W on Green 0 0 0.596 0.403
N on Goodwin
0 0.640 0.359 0
S on Goodwin
0 0.513 0 0.486
![Page 45: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/45.jpg)
45
Evaluation: Population Statistics
OptX: report last sampled location using sensor X (offline)
EnLoc-Deviate: Equally spaced GPS + population statistics (online). ALE ~ 32m
OptX: report last sampled location using sensor X (offline)
EnLoc-Deviate: Equally spaced GPS + population statistics (online). ALE ~ 32m
B = 25% Battery
![Page 46: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/46.jpg)
46
Future Work/Limitations
Assumed phone speed known Infer speed using accelerometer Energy consumption of accelerometer relatively small
Deviations from habitual paths Quickly detect/switch to deviation mode
Probability Map hard to build on wider scale Statistics from transportation departments
![Page 47: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/47.jpg)
47
Conclusions
Location is not for free Phone battery cannot be invested entirely into localization
Offline optimal accuracy computed For specified energy budget Known mobility trace
However, online localization technique necessary
EnLoc exploit prediction to reduce energy Personal Mobility Profiling Population Statistics
![Page 48: Energy-efficient Localization Via Personal Mobility Profiling Ionut Constandache Co-authors:](https://reader034.vdocument.in/reader034/viewer/2022051820/5681498e550346895db6d4c4/html5/thumbnails/48.jpg)
48
Questions?
Thank You!
Visit the SyNRG research group @http://synrg.ee.duke.edu/