/
1 April 14, 2011 1 April 14, 2011

1 April 14, 2011 - PowerPoint Presentation

faustina-dinatale
faustina-dinatale . @faustina-dinatale
Follow
409 views
Uploaded On 2016-07-18

1 April 14, 2011 - PPT Presentation

2 April 14 2011 Wheres WikiLeaks 3 April 14 2011 4 April 14 2011 5 April 14 2011 GEOLOCATION SERVICE Sándor Laki Eötvös Loránd University Budapest Hungary lakisinfeltehu ID: 409649

2011 april active geolocation april 2011 geolocation active service spotter delay normal distances ipligence databasesmaxmind infromationcommercial dnsorganizational basedwhois ip2location geolinguisticregistry passive general

Share:

Link:

Embed:

Download Presentation from below link

Download Presentation The PPT/PDF document "1 April 14, 2011" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.


Presentation Transcript

Slide1

1

April 14, 2011Slide2

2

April 14, 2011

Where’s

WikiLeaks

?Slide3

3

April 14, 2011Slide4

4

April 14, 2011Slide5

5

April 14, 2011Slide6

GEOLOCATION SERVICE

Sándor Laki*

Eötvös Loránd University

Budapest,

Hungary

lakis@inf.elte.hu

Spotter

:

A Model Based Active

Geolocation

Service

S. Laki*

, P.

Mátray

, P. Hága, T. Sebők, I. Csabai, G.

Vattay

The Project

was

supported by the European Union and co-financed by the European Social Fund (grant agreement no. TAMOP 4.2.1./B-09/1/KMR-2010-0003).Slide7

Motivation

7

April 14, 2011

Location information can be useful for both private and corporate users

Targeted advertising on the web

Restricted content delivery

Location-based security check

Scientific applications

Measurement visualization

Network diagnostics

Analysing

spatial properties

of

the

IntenetSlide8

IP

Geolocation

in

general

Passive GeolocationGeolinguistic

Registry

based

Whois

,

dns

Organizational infromationCommercial databases

Maxmind

,

IPLigence

, IP2Location, etc.

Active

Geolocation

8

April 14, 2011

rt1.lon.uk.geant2.net

London, UKSlide9

Geolinguistic

Registry

based

Whois, dns

Organizational infromationCommercial databasesMaxmind, IPLigence, IP2Location, etc.

rt1.lon.uk.geant2.net

IP

Geolocation

in

general

Passive

Geolocation

Active

Geolocation

9

April 14, 2011

Large

and

geographically

dispersed

IP

blocks

can

be

allocated

to

a

single

entitySlide10

Geolinguistic

Registry

based

Whois, dns

Organizational infromationCommercial databasesMaxmind, IPLigence, IP2Location, etc.

rt1.lon.uk.geant2.net

IP

Geolocation

in

general

Passive

Geolocation

Active

Geolocation

Active

probing

Delay

,

topology

, etc.

Landmarks

With

known

location

Location

estimates

for

each

individual IP addresses10April 14, 2011Large and geographically dispersed IP blocks

can be allocated to a single entitySlide11

Active

geolocation

11

April 14, 2011

?

Target

to

be

localizedSlide12

Active

geolocation

12

April 14, 2011

Landmarks

with

known

locationSlide13

Active

geolocation

13

April 14, 2011

200

ms

1010

ms

898

ms

501

msSlide14

Active

geolocation

14

April 14, 2011

Transforming

delays

to

geographical

distancesSlide15

15

April 14, 2011

How

to

transform

network

delays

to

geographical

distances

?Slide16

Our

delay–

distance model

Reference

data setNodes with known location

~700

PlanetLab

nodes

All

to

all

RTT

measurements

mininmum

delays

16

April 14, 2011Slide17

Our

delay-distance

model

Standardized values, fitted

normal distributionApproximated by a standard normal distributionGeographic distances follow

normal

distribution

with

parameters µ(t) and (t) for a given delay

t

17

April 14, 2011Slide18

Evaluation

18

April 14, 2011

The distances are normally distributed for a given RTT

For

each

RTT

the

radial

profil

e is

approximated

by

a

normal

distribution

(

m(

d

)

and

s(

d

)

)

=> which defines the spatial probability distribution of the targetSlide19

Evaluation

– „

Triangulation

Landmarks: L1, … , LnDelay di

from

L

i

to T: Li di

The

probability

that

target

T

resides

in

region

H

:

19

April 14, 2011Slide20

Performance

analysis

Estimated accuracy

for a geolocation

ground truthCAIDA’s Geolocation Comparision SurveyMore than

20000

reference

nodes

Located

in North America and in Europe

20

April 14, 2011

In

North

America

35%

9%

2

%

70

%

40

%

27

%Slide21

The Online service

21

April 14, 2011Slide22

Sp

otter

- The geolocation service

online: free, easy to use

~100 landmarksReal time measurements in the background

offline: batch mode

~5

0

k addresses/day

Full

landmark set ~700 nodesto fix:

DNS resolve problem

Sequential

queuing

22

April 14, 2011

visit

and

use

!

http://spotter.etomic.orgSlide23

23

April 14, 2011

Where’s

WikiLeaks

?Slide24

Conclusion

Spotter

is a measurement

based geolocation service

available onlineBased on a probabilistic approachA novel description of

delay-distance

relation

Validated

on

CAIDA’s Geolocation Ground Truth

30 km and 75 km

median

accuracy

in

North

America

and

in

Europe,

resp

.

New

challenges

:

IPv6, Real

time

measurements

,

etc

The Internet topology and the delay characteristics could be changed in

the future24April 14, 2011Slide25

25

April 14, 2011

Thank

you

for

your

attention

!

For

more

info

please

visit

http://spotter.etomic.org

or

send

me

an email

lakis

@

inf.elte.hu

(Sándor Laki)

GEOLOCATION SERVICE