Surveillance Sensor Networks

mashpeemoveΚινητά – Ασύρματες Τεχνολογίες

24 Νοε 2013 (πριν από 3 χρόνια και 8 μήνες)

166 εμφανίσεις

eTrack
: Target
Localization
System in
Surveillance
Sensor Networks

Graduate Research Symposium

May 04, 2012


Cuong

(Charlie) Pham

Challenges


Accuracy


Time


Cost


Noisy environment


Anchor locations need to be known in advance


Main Contributions:

-

Low cost

-
Working with noisy environment

-
Anchor locations unknown


Can we do better than this?

Of course, we do (APIT, Spotlight, Diffusion, etc.)

Equipment

Arduino

Uno SMD


-
ATmega328 microcontroller

-
32k Flash Memory

-
16Mhz Clock Speed

Xbee

Series 1 (802.15.4)

-
250kbps Max data rate

-
300ft (100m) range

Base Station

-

Network sink

The Machine Learning
Approach


Do classification to get location

o
Define classes

o
Get training data

o
Build model

o
Predict location





Training

Data

Model

Learning

Data

Class

Classes + Training Data

1

Class
1

Class 2

2

3

4

5

6

Location Estimation

Xi

Xi+1

Yj

Yj+1

The sensor is


In class Xi+1 but not in Xi


In class Yj+1 but not in
Yj

Binary Search

Xh

6

Accuracy

6

Demo Video


http://www.youtube.com/watch?v=BA6hUwSmWQ8

References

[1] XUANLONG
NGUYEN, MICHAEL I. JORDAN, and BRUNO
SINOPOLI.

A Kernel
-
Based Learning
Approach to
Ad Hoc
Sensor Network
Localization


[2]
Duc

A. Tran

and
Thinh

Nguyen.
Localization in Wireless Sensor Networks based on Support
Vector Machines.

IEEE Transactions on Parallel and Distributed Systems (TDPS), 19(7): 981
-
994,
July 2008.


[3] Wang, J.,
Ghosh
, R., and Das, S.
A survey on sensor localization
. Journal of Control Theory
and

Applications 8, 1 (2010), 2
-
11.


[4]
Lingxuan

Hu and David Evans. Localization for Mobile Sensor Networks. In Tenth Annual
International Conference on Mobile Computing and Networking (
MobiCom

2004).
Philadelphia, 26 September
-

1 October 2004


[5]
Tian

He, Chengdu Huang, Brian M. Blum, John A.
Stankovic
,
Tarek

Abdelzaher
. Range
-
Free
Localization Schemes for Large Scale Sensor Networks.

Come visit
NISLab

(S
-
3
-
124B)