Indoor Localization Without the Pain
description
Transcript of Indoor Localization Without the Pain
![Page 1: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/1.jpg)
Indoor Localization Without the Pain
Krishna ChintalapudiAnand Padmanabha Iyer
Venkata N. Padmanabhan
——presented by Xu Jia-xing
![Page 2: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/2.jpg)
Motivation Main idea of EZ Optimization Experiment Conclusion
Outline
![Page 3: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/3.jpg)
Motivation Main idea of EZ Optimization Experiment Conclusion
Outline
![Page 4: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/4.jpg)
Schemes that require specialized infrastructure. requires infrastructure deployment
Schemes that build RF signal maps. takes too much time
Model-Based Techniques. much less efforts than RF map; but still need a
lot of work to fit the models
Motivation-Related Work(1)
![Page 5: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/5.jpg)
Localization in Indoor Robotics. requires special sensors and maps
Ad-Hoc localization. requires enough node density to enable multi-
hopping
Motivation-Related Work(2)
Can we do indoor localization without such pre-deployments
or limitations?
![Page 6: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/6.jpg)
Works with existing WiFi infrastructure only
Does not require knowledge of Aps(placement, power,etc)
Even work with measurements by a single device
Does not require any explicit user participation
Motivation-EZ(1)
![Page 7: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/7.jpg)
There are enough WiFi APs to provide excellent coverage throughout the indoor environment
Users carry mobile devices, such as smartphones and netbooks, equipped with WiFi
Occasionally a mobile device obtains an absolute location fix, say by obtaining a GPS lock at the edges of the indoor environment, such as at the entrance or near a window.
Motivation-EZ(2)
Assumptions
![Page 8: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/8.jpg)
Motivation Main idea of EZ Optimization Experiment Conclusion
Outline
![Page 9: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/9.jpg)
Main idea of EZ-LDPL equations
![Page 10: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/10.jpg)
xj: the jth location ci: the ith AP’s location Pi: the power of the ith AP pij: the RSS received by mobile in the jth
location form the ith AP ri: the rate of fall of RSS in the vicinity of the
ith AP
Main idea of EZ
![Page 11: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/11.jpg)
Motivation Main idea of EZ Optimization Experiment Conclusion
Outline
![Page 12: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/12.jpg)
10% of the solutions with the highest fitness are retained.
10% of the solutions are randomly generated. 60% of the solutions are generated by crossover.
The remaining 20% solutions are generated by randomly picking a solution from the previous generation and perturbing it(Only Pi and ri)
Optimization-GA
Manner
![Page 13: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/13.jpg)
Randomly pick Pi and ri with boundaries
Use the LDPL equation :if there are m APs and n locationsthen reduce from 4m+2n to 4m
Optimization-Reducing the Search Space
![Page 14: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/14.jpg)
R1 : If an AP can be seen from five or more fixed (or determined)locations, then all four of its parameters can be uniquely solved.
R2 : If an AP can be seen from four fixed locations, there exist only two possible solutions for the four parameters of the AP.
R3 : If an AP can be seen from three fixed locations, randomly pick ri, there exist only two possible solutions for the three parameters of the AP.
Optimization-Reducing the Search Space
![Page 15: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/15.jpg)
R4 : If an AP can be seen from two fixed locations, randomly pick Pi and ri, there exist only two possible solutions for the two parameters of the AP.
R5 : If an AP can be seen from one fixed location, randomly pick all parameters.
R6 : If the parameters for three (or more) APs have been fixed, then all unknown locations that see all these APs can be exactly determined using trilateration.
Optimization-Reducing the Search Space
![Page 16: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/16.jpg)
Calculate all equations fit R1
Randomly generate parameters of all equations fit R2 to R5
Calculate parameters of all unknown locations
Optimization-Reducing the Search Space
![Page 17: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/17.jpg)
There are gain differences among different device.
Introduce an additional unkown parameter G
Optimization-Relative Gain Estimation Algorithm
![Page 18: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/18.jpg)
Calculate △Gk1k2 is possible:◦ represent all RSS from a device with a vector
Optimization-Relative Gain Estimation Algorithm
If “Close”
![Page 19: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/19.jpg)
Optimization-APSelect algorithm
Common Methods APSelect algorithm
Wide coverage
Low standard deviation in RSS
High average signal strength
Select each AP to provide information that other selected AP do not
1.Normalize pij into range(0,1)
2.Let
3.Cluster APs one by one by 入4.Select the AP which can be seen by most known locations.
![Page 20: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/20.jpg)
Motivation Main idea of EZ Optimization Experiment Conclusion
Outline
![Page 21: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/21.jpg)
Experiment-Performance
![Page 22: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/22.jpg)
Experiment-Performance
Normal accuracy.
![Page 23: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/23.jpg)
Experiment-Training Data
More training data greater accuracy.
![Page 24: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/24.jpg)
Experiment-new mobile
Great performance. Different devices are better.
![Page 25: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/25.jpg)
Experiment-Multiple devices training
The same as one device.
![Page 26: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/26.jpg)
Experiment-APSelect and LocSelect
Great improvement.
![Page 27: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/27.jpg)
Motivation Main idea of EZ Optimization Experiment Conclusion
Outline
![Page 28: Indoor Localization Without the Pain](https://reader030.fdocuments.in/reader030/viewer/2022033105/568165f7550346895dd91e94/html5/thumbnails/28.jpg)
The idea is good. It’s different from traditional methods.
The optimization is functional.
The LDPL Model is not perfect. Does not mention how to refresh the RSS
Model.
Conclusion