ISSN: 2278 – 1323

International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)

Volume 2, Issue 3, March 2013

Multiple Event-Driven Node Localization in

Wireless Sensor Networks
Martin Victor K, Mrs. K. Ramalakshmi

Abstract– In the area of wireless sensor networks,

localization is the essential issue. Localization - discovering the II. LOCALIZATION METHOD
location of individual sensor nodes leftovers one of the most
complex research challenges. Practical solutions involving Localization is an important issue in the WSN
reasonable power, computation and monetary costs do not technology. Localization problem refers to the method of
exist. Since many rising applications based on networked estimating and calculating the locations of sensor nodes.
sensors require location awareness, a node must be able to find The significance of this fact directed researchers to look for
its location. So, Spotlight, a localization system that carries a solution for localization problem. One simple way is
high-location estimation accuracy at low cost. The system uses
manual design but this is not viable in large scale or when
spatiotemporal properties of well-controlled events in the
network; light in this case, to obtain locations of sensor nodes. sensors are deployed in unreachable areas such as volcanoes
The system is to detect the multiple events in the network and or when sensors are movable. Another way is to add global
to increase the area of the sensor field by increasing the positioning system (GPS) to each sensor. GPS has affected
number of nodes. By handling this kind of detection of multiple by heavy trees and buildings because it requires line-of-
events in the network at once, mainly the time is saved. sight between the receiver and satellites. So, it has low
Spotlight offers various techniques that allow users to balance accuracy due to poor signal reception. As it is mentioned
time and accuracy to obtain results tailored to requirements. earlier, localization means finding the position of the
individual sensor nodes. In general, the localization method
Index Terms–Wireless Sensor Networks; Node Localization; can be classified into four different categories:
Multiple Events.  Centralized vs Distributed
 Anchor-free vs Anchor-based
 Range-free vs Range-based
 Mobile vs Stationary
I. INTRODUCTION

A wireless sensor network (WSN) consists of spatially Localization

scattered independent sensors to examine corporal or Methods

environmental circumstances, such as temperature, sound,

pressure, etc. and to considerately forward their data Centralized vs Anchor-free vs Range-free vs Mobile vs
Distributed Anchor-based Range-based Stationary
through the network. The WSN is formed of nodes - from a
very little to many, where every node is linked to one or
numerous sensors. Fig. 1 – Classification of Localization Method
Localization plays an important role. Here, Localization
problem means the method of approximating and figuring
out the positions of sensor nodes. The significance of this
III. EXISTING METHOD
information directed the researchers to find a solution for it.
One way is the manual design configuration but this is very Asymmetric Event-Driven Node Localization in Wireless
unworkable in wide-ranging or when sensors are deployed Sensor Networks proposes Spotlight, a localization system
in remote areas or when sensors are movable. Another way that gives high-location estimation accuracy at low cost.
is adding the global positioning system (GPS) to each Using an asymmetric design with all complicated hardware
sensor. Since GPS needs line-of-sight between the receiver and computation in a single device, Spotlight delivers a
and satellites, it has exaggerated heavy trees and big variety of methods that allow users to balance time and
buildings. Due to poor signal reception it has low accuracy. precision to attain outcomes modified to requirements. In all
cases, the only limiting aspect is the total size of the sensor
field. Any number of sensors may be identified within an
Martin Victor K, PG Scholar, Department of CSE, Karunya University, enclosed area at no additional cost, making Spotlight
Coimbatore, India.
Mrs. K. Ramalakshmi, Asst. Professor, Department of CSE, Karunya suitable for large-scale deployments.
University, Coimbatore, India. The basic concept of the Spotlight localization system is
the generation of controlled events detectable by deployed

1073
All Rights Reserved © 2013 IJARCET
 ISSN: 2278 – 1323
International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 2, Issue 3, March 2013

sensor nodes. Events like light and sound, with well functionality of the Spotlight localization system. Each of
characterized spatiotemporal properties and detectable with these designs is evaluated in three scenarios: 1) the terrain is
simple sensing hardware, perform well in this system. By known or assumed to be flat; 2) terrain information is
determining a sensor node’s recognition time of a generated available; and 3) the terrain is unknown.
event, a spatial relationship between the sensor node and the Point Scan, Line Scan, and Area Cover EDFs all localize
event generator can be inferred. There are three main sensor nodes. However, they vary in localization time,
functions: Event Detection, Event Distribution Function and communication overhead, and energy consumed (defined as
Localization Function. Event Overhead). Assume that all sensor nodes are located
The EDF E(t) may be adjusted to disseminate events in a square with edge size D, and that the Spotlight device
optimally based on limitations forced by sensor abilities, can generate N events (e.g., Point, Line, and Area Cover
limitations of the platform transporting the Spotlight system, events) every second, and that the maximum tolerable
limitations forced by terrain, and availability of detailed localization error is r. Table 1 compares the execution cost
geographic information. The Point Scan, Line Scan, and of the three techniques.
Area Cover Event Functions each illustrate basic

Table I - Execution Cost Comparison Criterion

Criterion Point Scan Line Scan Area Cover
2 2
Localization Time (D /r )/N (2D/r)/N logrD/N
# Event Detections 1 2 logrD
2 2 2
Event Overhead D 2D D logrD/2

IV. PROPOSED METHOD AND ARCHITECTURE

Some of the limitations in the existing concept are as

follows:
 The limiting factor here is the total size of the
sensor field. Because of this the coverage is
minimized. To overcome this, the number of nodes
should be increased to increase the total size of the
sensor field.
 Here only single event is detected and the multiple
events are not worked out.
 High precision of localization is not up to the level Fig. 2 – Spotlight system architecture, depicting the multiple
with respect to the time and accuracy. events
To overcome these limitations, the following concepts are
enhanced. In the existing concept, single event is detected With the support of these three functions, the localization
and driven, but here, the multiple events detection concept is process proceeds as follows:
discussed and implemented. Addition to that, the coverage  A Spotlight device distributes events in the space
is maximized by increasing the number of nodes and also A.
the localization concept is précised with respect of time and  During event distribution, sensor nodes record the
accuracy. time sequence Ti = {ti1, ti2,…., tim} at which they
Any number of sensors may be localized within a covered detect the events.
area at no additional cost, making Spotlight suitable for  After event distribution, each sensor node sends the
large-scale deployments. Events like light and sound, with detection time sequence Ti to the Spotlight device.
well characterized spatiotemporal properties and detectable  The Spotlight device estimates the location of a
with simple sensing hardware, perform well in this system. sensor node ni, using Ti and the known E(t)
In further, there are three main functions which support function.
the above said modules in the implementation and they are Event Detection D(e) is carried by the sensor nodes. It
as follows: the Event Distribution Function (EDF) E(t), the resolves whether a peripheral event occurs or not.
Event Detection Function D(e), and the Localization Localization Functions L(Ti) are applied by Spotlight
Function L(Ti). devices Event Distribution E(t) depicts the dispersal of

1074
All Rights Reserved © 2013 IJARCET
 ISSN: 2278 – 1323
International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 2, Issue 3, March 2013

events over time and is present on one or more Spotlight Compute Event Detection Func_ D(e) (Node Id, Event Id,
devices. time Ti)
The EDF is given by For each sensor node
𝑃 𝑃 𝑥, 𝑦 ∈ 𝐴 𝑎𝑛𝑑 𝑥 = 𝑠𝑡 𝑚𝑜𝑑 𝑙 Do
E(t) =
𝑎𝑛𝑑 Send detection time sequence T i = {ti1, ti2,…., tim}
y = 𝑠𝑡/𝑙 𝑟}. to spotlight device
The resulting localization function is End
L(Ti) = E(Ti1) = {(sti1)mod(l), sti1/𝑙 𝑟} Compute Localization function_L(t) ( sensor node Id, time
There are two terrains: known terrain and unknown Sequence)
terrain. Now, the equation of the line from Spotlight device For each node do
to the terrain is given by Estimate location of node ni,
𝑥 𝑥1 cos 𝛼1 Process with Ti and E(t)
𝑦 = 𝑦2 + k cos 𝛽1 End
𝑧 𝑧3 cos 𝛾1 End Proc_ Localization
where k is given by
𝑥−𝑥1 𝑦 −𝑦1 𝑧−𝑧1
k= = =
cos 𝛼1 cos 𝛽1 cos 𝛾1 B. Flow Chart
and
𝑥 𝑧−𝑧1 cos 𝛼1 𝑥1 Spotlight Device
𝑦 = cos 𝛾1 cos 𝛽1 + 𝑦1
Now, the resulting localization function is Event Distribution
𝑥 𝑧−𝑧1 cos 𝛼1
L(Ti) = E(Ti1) = max 𝑦 = +
cos 𝛾1 cos 𝛽1

𝑥1𝑦1 𝑎𝑛𝑑 𝑥𝑦𝑧 ∈𝐻𝑀

Sensor Node
where 𝛼1, 𝛽 1, and 𝛾1 are functions of time and are calculated
using the time stamp ti1 received. Event Detection – records the
time sequence and sends
V. SYSTEM IMPLEMENTATION

The implementation of this project is planned to be done

with following modules. These modules would help in Spotlight Device
practical implementation of this project. Each module would
be made through precise steps towards the successful Localization Function – receives
implementation. The modules are as follows:
the time sequence
 Increasing the total size of the sensor field by
maximizing the coverage and increasing the
number of nodes.
 Distribution and detection of multiple events.
 High localization precision with respect to the time Location Estimation
and accuracy.
Sensor Node
A. Algorithm
Start Proc_ Localization (Event_Id, Source Address,
Destination Address, Device ID)
VI. PERFORMANCE EVALUATIONS
Assume a space A ∁ R3 and a spotlight device
Initialize Sensor nodes For discussing the results, there are four parameters used
For each node n to compare the single event detection with the multiple
Do events detection, they are as follow:
Position node at Pi(xi, yi, zi)  Nodes vs Packet Delivery Ratio
Compute Event Distribution Func_EDF( Event Id, Node Id,  Nodes vs Delay
Address)  Nodes vs Overhead
Record the time Sequence Ti = {ti1, ti2,…., tim}
 Nodes vs Throughput
End

1075
All Rights Reserved © 2013 IJARCET
 ISSN: 2278 – 1323
International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 2, Issue 3, March 2013

Fig. 3(a) – Nodes vs Packet Delivery Ratio, Fig. 3(b) – Nodes vs Delay, Fig. 3(c) – Nodes vs Overhead, Fig. 3(d) –
Nodes vs Throughput

The fig. 3(a) depicts the graph comparison between the events detection. The nodes lie through the x-axis and the
single event and multiple events detection with respect of throughput lies through the y-axis. In this graph, there is a
nodes and the packet delivery ratio. Here, the red line shows variation occurs in the multiple events detection when
the single event detection and the green line shows the compared with the single event detection.
multiple events detection. The nodes lie through the x-axis
and the packet delivery ratio lies through the y-axis.
The fig. 3(b) depicts the graph comparison between the VII. CONCLUSION
single event and multiple events detection with respect of Based on the limitations of the existing model, it is
nodes and the delay. Here, the red line shows the single decided to implement three enhancement ideas: To
event detection and the green line shows the multiple events maximize the coverage by increasing the total size of the
detection. The nodes lie through the x-axis and the delay lies sensor field and also by increasing the number of nodes; To
through the y-axis. In this graph, there is a variation occurs distribute and detect the multiple events; High Localization
in the multiple events detection when compared with the precision with respect to time and accuracy. Localization is
single event detection. finding the position of individual sensor nodes remains one
The fig. 3(c) depicts the graph comparison between the of the most difficult research challenges. Here, this method
single event and multiple events detection with respect of presents the design, implementation, and evaluation of
nodes and the overhead. Here, the red line shows the single Spotlight, a localization system for wireless sensor
event detection and the green line shows the multiple events networks. The system is divided into different modules and
detection. The nodes lie through the x-axis and the overhead each module is processed separately and finally
lies through the y-axis. In this graph, there is a variation implemented together.
occurs in the multiple events detection when compared with
the single event detection.
The fig. 3(d) depicts the graph comparison between the
single event and multiple events detection with respect of
nodes and the throughput. Here, the red line shows the
single event detection and the green line shows the multiple

1076
All Rights Reserved © 2013 IJARCET
 ISSN: 2278 – 1323
International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 2, Issue 3, March 2013

REFERENCES Mrs. K. Ramalakshmi received her

bachelor of engineering degree in computer
[1] “Radar: An in-Building Rf-Based User Location and science in the year 2002. She received her
Tracking System,” Proc. IEEE INFOCOM, 2000 Bahl and master of engineering in the year 2006. She
Padmanabhan. is currently working as an Assistant
[2] K. Lorincz and Welsh, “Motetrack: A Robust Decentralized Professor at Karunya University,
Approach to RF-Based Location Tracking,” Proc. Int’l Workshop Coimbatore, India. She has a teaching
Location and Context-Awareness, 2005. experience of over 7 years. Also she is
[3] Corke. P, Peterson. R, and Rus. D, “Networked Robots: Flying Robot currently pursuing her research work in the
Navigation Using a Sensor Net,” Proc. Int’l Symp. Robotics Research area of wireless sensor networks.
(ISRR), 2003.
[4] “Localization for Mobile Sensor Networks,” Proc. ACM MobiCom,
2004 L. Hu and D. Evans.
[5] Pathirana, Savkin, Jha, and Bulusu, “Node Localization Using Mobile
Robots in Delay-Tolerant Sensor Networks,” IEEE Trans. Mobile
Computing, vol. 4, no. 3, pp. 285-296, May/June 2005.
[6] “Localization of Wireless Sensor Networks with a Mobile Beacon,”
Proc. IEEE Int’l Conf. Mobile Ad-Hoc and Sensor Systems, 2004 M.
Sichitiu and V. Ramadurai.
[7] Priyantha, Chakraborty, and Balakrishnan, “The Cricket Location-
Support System,” Proc. ACM MobiCom, 2000.
[8] “Dynamic Fine-Grained Localization in Ad-Hoc Networks of
Sensors,” Proc. ACM MobiCom, 2001 A.Savvides, C. Han, and M.
Srivastava.
[9] “Sensor Network-Based Countersniper System,” 2004, Simon,
Maroti, Ledeczi, Balogh, Kusy, Nadas, Pap, Sallai, and Frampton.
[10] Barton-Sweeney, Lymberopoulos, and Savvides, “Sensor
Localization and Camera Calibration in Distributed Camera Sensor
Networks,” Proc. Third Int’l Conf. Broadband Comm.,Networks and
Systems (BROADNETS), pp. 1-10, 2006.
[11] “GPS-Less Low Cost Outdoor Localization for Very Small Devices,”
IEEE Personal Comm. Magazine, vol. 7, no. 5, pp. 28-34, Oct. 2000,
N. Bulusu, J. Heidemann, and D. Estrin.
[12] Schuhmann, Herrmann, Rothermel, Blumenthal, and D.Timmermann,
“Improved Weighted Centroid Localization in Smart Ubiquitous
Environments,” 2008.
[13] He, Huang, Blum, Stankovic, and Abdelzaher, “Range-Free
Localization Schemes in Large Scale Sensor Networks,” Proc. ACM
MobiCom, 2003.
[14] “Organizing a Global Coordinate System from Local Information on
an Ad Hoc Sensor Network,” Proc.Second Int’l Conf. Information
Processing in Sensor Networks (IPSN), 2003, R. Nagpal, H. Shrobe,
and J. Bachrach.
[15] Niculescu and B. Nath, “Ad-Hoc Positioning System,” Proc. IEEE
Global Telecomm. Conf. (GLOBECOM), 2001.
[16] “MSP: Multi-Sequence Positioning of Wireless Sensor Nodes,” Proc.
Int’l Conf. Embedded Networked Sensor Systems (SenSys), 2007,
Zhong and T. He.
[17] Zhong, D. Wang, and He, “Sensor Node Localization Using
Uncontrolled Events,” Proc. Int’l Conf. Distributed Computing
Systems (ICDCS), 2008.
[18] “A High-Accuracy Low-Cost Localization System for Wireless
Sensor Networks,” Proc. Int’l Conf. Embedded Networked Sensor
Systems (ENSS), 2005, R. Stoleru, T. He, J.A. Stankovic, and D.
Luebke.

Martin Victor K received the bachelor of

engineering degree in computer science from
CSI College of Engineering, Ketti – The
Nilgiris under Anna University, Chennai,
India, in 2006. He is currently pursuing the
master of technology degree in software
engineering at Karunya University,
Coimbatore, India. His research interests
include Computer Networks, Internet and
distributed systems.

1077
All Rights Reserved © 2013 IJARCET