Localization in WSN

Localization in WSN 1

Localization in WSN

Presented by: Yara AliSupervised by: Dr. Ahmed

Akl


Agenda Introduction to WSN Localization Usage GPS .. Why not ? Localization methods taxonomy Classifications of Localization Methods Summary Future work References


Introduction to WSN A large number of self-sufficient nodes

Nodes have sensing capabilities

Can perform simple computations

Can communicate with each other


Introduction to WSN (Cont.)

Beacon (Anchor) node:

It’s a node that’s aware of it’s location, either through GPS or manual pre-programming during deployment.


Introduction to WSN (Cont.) In a Wireless sensor nodes thousands of

sensors need to know their position

Many applications need position info: in-home forest-fire detection atmospheric (temperature,

pressure, … ) military (target detection, …) police



Advantages:1. It avoids a lot of wiring 2. It can accommodate new devices

at any time 3. It's flexible to go through physical

partitions 4. It can be accessed through a

centralized monitor



Disadvantages 1. It's easy for hackers to hack it as

we cant control propagation of waves

2. Comparatively low speed of communication

3. Gets distracted by various elements like Blue-tooth


Localization Localization is a process to compute the

locations of wireless devices in a network

WSN Composed of a large number of inexpensive nodes that are densely deployed in a region of interests to measure certain phenomenon.

The primary objective is to determine the location of the target


Localization (CONT.)


Localization (CONT.)


Usage

Coverage Deployment Routing Location service Target tracking rescue


GPS .. Why not ? We need to determine the physical

coordinates of a group of sensor nodes in a wireless sensor network (WSN)

Due to application context and massive scale, use of GPS is unrealistic, therefore, sensors need to self-organize a coordinate system


GPS .. Why not ? (Cont.)1. Expensive

2. GPS satellite signals are weak (when compared to, say, cellular phone signals), so it doesn't work as well indoors, underwater, under trees, etc.

3. The highest accuracy requires line-of-sight from the receiver to the satellite, this is why GPS doesn't work very well in an urban environment

4. The US DoD (dept of defense) can, at any given time, deny users use of the system (i.e. they degrade/shut down the satellites)


Localization methods taxonomy


1- Target/Source Localization

Most of the source localization methods are focused on the measured signal strength.

To obtain the measurements, the node needs complex calculating process.


1- Target/Source Localization (Cont.)

1. The received signal strength of single target/source localization in WSN during time interval t:


1- Target/Source Localization (Cont.)

2. The received signal strength of multiple target/source localization in WSN during time interval t:


1- Target/Source Localization (Cont.) The Above methods require transmission

of a large amount of data from sensors which may not be feasible under communication constraints.

3-4. The binary sensors sense signals ( infrared, acoustic, light, etc. ) from their vicinity, and they only become active by transmitting a signal if the strength of the sensed signal is above a certain threshold.


1- Target/Source Localization (Cont.) The binary sensor only makes a binary

decision (detection or non-detection) regarding the measurement.

Consequently, only its ID needs to be sent to the fusion center when it detects the target. Otherwise, it remains silent.

So, the binary sensor is a low-power and bandwidth-efficient solution for WSN.


Taxonomy


2- Node Self-localization Range-based Localization: uses the

measured distance/angle to estimate the indoor location using geometric principles.

Range-free Localization: uses the connectivity or pattern matching method to estimate the location. Distances are not measured directly but hop counts are used. Once hop counts are determined, distances between nodes are estimated using an average distance per hop and then geometric principles are used to compute location.


2-1 Range based localization


2-1 Range based localization (Cont.)

1. Time of arrival: (TOA)

It’s a method that tries to estimate distance between 2 nodes using time based measures.

Accurate but needs synchronization



2. Time Difference Of Arrival: (TDOA) It’s a method for

determining the distance between a mobile station and a nearby synchronized base station. (Like AT&T)

No synchronization needed but costly.



3. Received Signal Strength Indicator: (RSSI)

Techniques to translate signal strength into distance

Low cost but very sensitive to noise



4. Angle Of Arrival: (AOA) It’s a method that allows

each sensor to evaluate the relative angles between received radio signals.

Costly and needs extensive signal processing.


2-2 Range-free localization

DV-Hop is the typical representation

It doesn’t need to measure the absolute distance between the beacon node and unknown node. It uses the average hop distance to approximate the actual distances and reduces the hardware requirements.


2-2 Range-free localization (Cont.)

Adv:Easy to implement and applicable to large network.

Disadv:The positioning error is correspondingly increased.


2-2-1 DV-Hop

It is divided into 3 stages:

1. Information broadcast2. Distance calculation3. Position estimation


1-Information broadcast The beacon nodes broadcast their location

information package which includes hop count and is initialized to zero for their neighbors.

The receiver records the minimal hop of each beacon nodes and ignores the larger hop for the same beacon nodes.

The receiver increases the hop count by 1 and transmits it to neighbor nodes.

All the nodes in a network can record the minimal hop counts of each beacon nodes.


2-Distance calculation According to the position of the beacon

node and hop count, each beacon node uses the following equation to estimate the actual distance of every hop


3- Position estimation The beacon node will calculate the

average distance and broadcast the information to network.

The unknown nodes only record the first average distance and then transmit it to neighbor nodes.

The unknown node calculates its location through.


2-2-1 DV-Hop (Cont.) A-B: 15

Anchors flood network with

own position flood network with avg hop distance

Nodes count number

of hops to anchors multiply with avg hop distance

A

B

1

1

1

1

2

2

2

3

3

4

4

3 hopsavg hop: 5

C


2-2-1 Modified DV-Hop


2-2-2 Pattern Matching Localization Also called map-based or finger print

algorithm.

It involves 2 phases:1. The received signals at selected locations are

recorded in an offline database called radio map.

2. It works at the online state.

The pattern matching algorithms are used to infer the location of unknown node by matching the current observed signal features to the prerecorded values on the map


Classifications of Localization Methods The localization techniques can be classified

with respect to various criteria:

1. Centralized vs Distributed

2. Range-free vs Range-based

3. Mobile vs Stationary

4. Coarse-grained vs fine-grained


Centralized vs Distributed Centralized

Data collected in the whole network are transmitted to the central unit that calculates the estimated location of each node in a network.

Distributed Computation is distributed among the nodes Each node estimates its own position based

on the local data gathered from its neighbors.


Range-Free vs Range-Based Range-Free (connectivity)

Makes no assumption about the availability or validity of such information, and use only connectivity information to locate the entire sensor network.

Hop-Counting Techniques

Range-Based (distance) Defined by protocols that use absolute point to

point distance estimates (range) or angle estimates in location calculation.


Mobile vs Stationary Mobile Stationary


Coarse-grained vs fine-grained

Coarse-grained:finding approximate coordinates of nodes in a network so it provide lower precision estimates of this coordinates.

Fine-grained:Determining precisely the coordinates but require much more communication and computation efforts.


Summary WSN .. What & Why ?

Distance estimation VS position computation VS localization algorithm

Single/Multiple localization in WSN/WBSN

Calculating the distance between sensor nodes ( TOA – TDOA – RSSI – AOA )


Summary Range-based methods require extra hardware therefore

have a higher cost but provide more accurate distance measurements, whereas range-free methods use only connectivity information and so are less accurate.

Range-free localization ( DV-Hop , Modified DV-Hop , pattern matching localization )

The localization techniques can be classified with respect to various criteria. They differ on the assumed localization precision, hardware capabilities, measurement and calculation methods, computing organization, the assumed network configuration, architecture, nodes properties and deployment, etc.


Future Work

Few papers investigate multiple-source localization in WSN and WBSN


References

1. http://www.hindawi.com/journals/ijdsn/2012/962523/

2. http://www.docslide.com/wireless-sensor-networks-nodes-localization-issue/

3. http://www.docstoc.com/docs/32678966/Localization-in-Wireless-Sensor-Network---ELEC-619B-Presentation


References (Cont.)4.https://www.cs.virginia.edu/~stankovic/psfiles/

wsn.pdf

5.http://www.sersc.org/journals/IJCA/vol6_no3/7.pdf

6.http://www.docstoc.com/docs/130374399/Localization-in-Wireless-Sensor-Networks

7. http://www.degruyter.com/view/j/amcs.2012.22.issue-2/v10006-012-0021-x/v10006-012-0021-x.xml


Any Questions?


Thank You !

Education

Localization in WSN