network kernel architectures and implementation (01204423) localization chaiporn jaikaeo...

19
Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo [email protected] Department of Computer Engineering Kasetsart University Materials taken from lecture slides by Karl and Willig

Upload: richard-mosley

Post on 13-Jan-2016

217 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

Network Kernel Architectures

and Implementation(01204423)

Localization

Chaiporn [email protected]

Department of Computer EngineeringKasetsart University

Materials taken from lecture slides by Karl and Willig

Page 2: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

2

Overview Basic approaches Trilateration Multihop schemes

Page 3: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

3

Localization & positioning Determine physical position or

logical location Coordinate system or symbolic

reference Absolute or relative coordinates

Metrics Accuracy Precision Costs, energy consumption, …

http://www.mathsisfun.com/accuracy-precision.html

Page 4: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

4

Main Approaches Based on

information source Proximity (Tri-/

Multi-)lateration and angulation

Scene analysis Radio environment

has characteristic “signatures”

Length known

Angle 1

Angle 2

(x = 2, y = 1)

(x = 8, y = 2)

(x = 5, y = 4)

r1

r2

r3

Page 5: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

5

Estimating Distances – RSSI Compute distance from Received

Signal Strength Indicator

Problem: Highly error-prone process

Distance

PD

F

DistanceSignal strength

PD

F

Page 6: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

6

Estimating Distances – Others Time of arrival (ToA)

Use time of transmission, propagation speed, time of arrival to compute distance

Time Difference of Arrival (TDoA) Use two different signals with different

propagation speeds Example: ultrasound and radio signal

Page 7: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

7

Determining Angles Directional antennas

Multiple antennas Measure time difference between

receptions

Page 8: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

8

Range-Free Techniques Overlapping connectivity

Approximate point in triangle

?

?A

B

C

D

F

G

E

Page 9: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

9

Overview Basic approaches Trilateration Multihop schemes

Page 10: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

10

Trilateration Assuming distances to

three points with known location are exactly given

Solve system of equations(x1,y1)

(x2,y2)

(x3,y3)

(xu,yu)

r1r2

r3

Page 11: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

11

Trilateration as Matrix Equation Rewriting as a matrix equation:

Example: (x1, y1) = (2,1), (x2, y2) = (5,4), (x3, y3) = (8,2), r1 = 100.5 , r2 = 2, r3 = 3

3

Page 12: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

12

Trilateration with Distance Errors What if only distance estimation ri' = ri + i

available? Use multiple anchors

Overdetermined system of equations

Use (xu, yu) that minimize mean square error, i.e,

Page 13: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

13

Minimize Mean Square Error

Look at derivative with respect to x, set it equal to 0:

Normal equation Has unique solution (if A has full rank),

which gives desired minimal mean square error

Page 14: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

Example: Distance Error Anchors' positions and measured

distances:

14

(x,y) r

(2,1) 5

(5,4) 1

(8,2) 4

(3,1) 2

(7,5) 3

(2,8) 7

(4,6) 4

7.2

5.5x̂

Solve bAxAA TT ˆ

0.5

Page 15: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

15

Overview Basic approaches Trilateration Multihop schemes

Page 16: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

16

Multihop Range Estimation No direct radio communication exists

Idea 1: Count number of hops, assume length of one hop is known (DV-Hop)

Idea 2: If range estimates between neighbors exist, use them Improve total length of route estimation

in previous method (DV-Distance)

X

B

A

C

Page 17: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

17

Iterative Multilateration

Page 18: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

18

Probabilistic Position Description Position of nodes is only probabilistically

known Represent this probability explicitly Use it to compute probabilities for further

nodes

Page 19: Network Kernel Architectures and Implementation (01204423) Localization Chaiporn Jaikaeo chaiporn.j@ku.ac.th Department of Computer Engineering Kasetsart

19

Conclusions Determining location or position is a

vitally important function in WSN, but fraught with many errors and shortcomings Range estimates often not sufficiently

accurate Many anchors are needed for

acceptable results Anchors might need external position

sources (GPS) Multilateration problematic

(convergence, accuracy)