Third International Conference on Devices, Circuits and Systems (ICDCS'16)

170

AnAdaptive localization error minimizationapproach

for wireless sensor network
Sudha H.Thimmaiah

Dr. Mahadevan.G.

Associate professor, Dept. of TelecommunicationEngg.

Dr. Ambedkar Institute of Technology, VTU
Bengaluru, India.
sudhacseait@rediffmail.com

Principal, Annai College of Engg.

Kumbakonam, India.
g_mahadevan@yahoo.com

AbstractWireless sensor based communication system isan

ever growing sector in the industry of communication. Wireless
infrastructure
is
a
network
that
enablescorrespondencebetweenvarious
devices
associated
through a infrastructure protocol. Finding the position or
location of sensor node (Localization) is an important factor in
sensor network for proving efficient service to end user. The
existing technique proposed so for suffers in estimating the
likelihood of localization error. To cater this in this work the
author proposes a RSS (Received signal strength) based
localization technique and also proposes an adaptive information
estimation to reduce or approximate the localization error in
wireless sensor network. The author compares our proposed
localization model with existing protocol and analyse its
efficiency.
KeywordsWsn, localization, radio channel measurement.

I.

INTRODUCTION

In wireless sensor network communications have enabled

low cost, power, dynamic sensor devices for the development
that are tiny and short communication range. The WSN used
tiny (small) sensor nodes, which consist of sensing, data
processing, and collaborating (communication) component. A
sensor device is a device in a WSN which perform the task like
rarefaction, sensing activity for data transmission and
interactiveamong sensor devices in the wireless sensor
infrastructure.
We are living in an environment loaded with sensors
devices. The structures that we do our daily activity in have
sensors checking temperature, inhabitance, pollution and
safety. A WSNs has key/important application/apps that are
used for remote tracking ecological monitoring, detecting
wildfire or sign of any catastrophic phenomenon, intruder
detection in military application. This has been enabled by the
availability, especially in recent years, of sensors that are small
in size. These nodes are furnished with wireless interface (WI)
that enables them to connect through each other to form a
wireless infrastructure.
WSNs guarantee to enhance the correspondence of critical
data in various industries of uses. As of late, WSN-based
SOA(service oriented based architecture) has ended up
progressively key for building adaptable and effective
application
administrations
[1]
where
application
administrations live in distinctive layers of data frameworks

978-1-5090-2309-7/16/$31.002016 IEEE

[3]. Consistently incorporating application administrations into

WSNs is an interesting issue of administration based
applications. The general point of SOA-based WSN is to
progressively make or arrange administrations that can give
solid and persistent accessibility of administrations, which can
boost the advantages and diminish the expenses of utilizations
[1]. Numerous SOA-based applications, for example,
information storage and management have been explored in [3]
and [4]. In [5], a SOA is proposed for industrial purpose, where
WSNs are connected as an application administrator. SOA
based applications such as Monitoring ecological application,
GPS based application, also, medicinal services have been
considered services in [5][7]. In sensor network the
localization of node has turned into a crucial prerequisite for
providing effective service to end user. Many location based
service are provided to localize the node but it is practically
difficult to provide an effective localization technique.Recently
many localization techniques have been proposed based on
GPS (global positioning system)[10], [11] to solve the
localization problem as in [8][10] but these technique suffers
from energy efficiency issues and the computational cost is
also high and it is very expensive to localize the node due
topology dynamic as follows (large number of nodes,
equipping the sensor with GPS, etc..). An alternative approach
based on distance measurement of reference and unknown
node is performed by the following technique TDoA (Time
difference on arrival), ToA (time on arrival), RSS (Received
signal strength) and so on. The TOA-based localization system
[19] [20] requires additional hardware to guarantee
synchronization between a transmitting node and a receiving
node. More importantly, a little of time inconsistency will
cause a significant ranging error. Similarly, TDOA-based
localization systems share similar drawbacks with TOA
systems, because they all involve expensive equipment.
Additionally, TOA/TDOA methods generally require a highly
dense deployment of sensor nodes, due to their shorter signal
propagation distances. i.e the propagation distance of
ultrasound is approximately 1525 feet. Moreover, AOA [17]
[18] systems estimate the distance amongdevices by referring
to the relative angles measured through an angle measuring
device. Although AOA systems can achieve high localization
accuracy, they are not suitable for large-scaled or highly dense
WSNs due to their complicated architecture and high costs.
RSS-based localization techniques, on the other hand,
overcome these shortcomings mentioned above. The principle
of the RSS methods is to utilize some Radio Frequency (RF)

Third International Conference on Devices, Circuits and Systems (ICDCS'16)

signal propagation models derived from either theoretical
inference or empirical results to make a translation of RSS to
measure the distance without requiring additional ranging
equipment. Such methods are cost-effective and can be
conveniently implemented since most RF IC-chips used in
WSN nodes can directly provide the information of RSS.
Below is some of the challenges in wireless sensor network
that need to be consider in localization.
A. Issues and challenges faced in reduecing error
Communication among sensor devices: huge amount
packet transmissionin wireless sensor network is merely not
practicalcontrastingto other wireless infrastructure devicessuch
as router, network interface card, etc., since these sensor
devices are designed to cater low power requirement.
Networking: sensor networks which loaded with numerous
sensor devices often lead to volatile outcomedue to topology
dynamic as result it leads to an unbalance consumption of
energy, higher collision rates, and channel contention among
nodes.
Data confidentiality: To safeguard the obtained data
information from unauthorized user. To achieve this various
user authentication and cryptographic protocol is been
proposed for sensor network.
Management: sensor network require proficient protocol
strategy to extract their potential of following thing such power
management, radio device interference management, battery
management of sensor devices, hardware and software bug
management, node failure, data loss due node failure etc...
Minimize the error in distance measurement:The path loss
model can be determined, where the prototype parameters can
be evaluated based on obtained data by using measurement of
RSS [8][9], [12]. Such prototypefixes the distance from the
unknown node to at least few nodes with known locations.
Then, classical approaches may be employed to obtain the
position of the unknown node. The drawback of this kind of
model are low scalability, topology sensitivity and
unreliability, [12][13].
Accept the noisy measurement but perform a global
optimization to reduce the error in estimating thelocation [9]
[11], [17] and [18]. One well known methodology is based on
MDS (multidimensional scaling)[10], [13], [15]. These
methodology aids at finding a low-dimensional location
mapping from a high-dimensional distance measurement space
by reducing the pairwise detachment errors [14]. These
measured distances form a matrix that can be used to
reconstruct a 2-D or 3-D coordinate system by MDS. In this
scenario the processes such as shifting, reflection, or the
combination of both can be used to keep the distance among
the points invariable.

171

and the experimental study are presented in the section three.

The concluding remark is discussed in the last section.
II.

PROPOSED MODEL

Herewe specialize for device location estimation using pairwise RSS measurements in a wireless network. Specifically,
consider a network of referenceand blindfolded nodes. The
node parameters
where, for a two-dimensional
The relative location problem
system,
corresponds to the estimation of blindfolded device coordinates
.
(1)
given
coordinates

the

known

reference
.

In RSS
is the measured received power at device
transmitted by device (in milliwatts). The author consider
only subset of
of devices that make pair-wise
measurements with device ,
and
are
taken to be upper triangular matrices, and these measurements
are assumed statistically independent. In addition, assume
thatt
is Gaussian distributed with mean
, which is denoted as

and variance
(2)

Where is the speed of propagation, and is not a functionof

. We assume that
is log-normal; thus, the random
is Gaussian
variable

(3)
Where
is the average power in decibel
is the variance of the shadowing, and
milliwatts,
is the received power in decibel milliwatts at a reference
.Generally
and is calculated from the
distance
free space path loss formula [16]. The path loss exponent is
a
function
of
the
environment.For
particular
environments, may be known from prior measurements.
Although we derive the Adaptive information estimation
strategy assuming is known, it could have been handled as an
unknown parameter.
Given (3), the density off is

In this work the author propose a RSS (Received signal

strength) based localization technique and also proposes an
adaptive information estimation to reduce or approximate the
localization error in wireless sensor network.

(4)

The paper organization is as follows: The proposed

localizationmodels are presented in Section two. The results

Here
is the AIES of range given received power
Neither
nor
are assumed to be ergodic random
variables; in fact, obstructions in the measured environment

Third International Conference on Devices, Circuits and Systems (ICDCS'16)

that cause shadowingerrors do not usually change over time.
The AIES gives a lower bound on the ensemble variance over
different random shadowing environments. If networks with
the same relative device coordinates are implemented in many
different areas, the variances of any unbiased coordinate
estimator will be lower bounded by the AIES presented here.
A. Adaptive information estimation strategy(AIES)
In sensor infrastructure in order to solve the problem of
self-calibration, parameters of all nodes in ainfrastructure must
be determined. Dada info comes both from measurements
made between pairs of nodes and a subset of nodes that know a
priori their bounds. Ainfrastructure self-calibration estimator
calculated the unknown nodesbounds. Let consider, distributed
clock synchronization in ainfrastructure could be accomplished
by nodesperceiving pair-wise timing offsets when just a small
number of devices are synchronous.
Specifically,
consider
a
vector
of
nodebounds
each
device
has
one
parameter. Devices
are blindfolded devices, and devices
are reference devices. The unknown parameter
where
for
vector is
Note that
are known. Devices
with
density
make
pair-wise
observations
We allow for the case when devices make
incomplete observations since two devices may be out of range
device makes
or have limited link capacity. Let
pair-wise observations with device
The pdf joint conditional is defined as
(5)

172

algorithm and we have varied devices size by 2, 4, 6 and 10

and conducted simulation study, simulation parameters are
given in table 1. These devices (nodes) are placed in a random
manner in a simulation area of 100 X 100 which is two
dimensional. For each mobile anchor node radius of
transmission and area of transmission is defined. The
performance evaluation strategy is adopted by varying the
number of nodes, variations in obstructions and orientations.
Table.1. Simulation parameters
Network Area

100 X 100

Sensor Per Side

2,4,6

Reference Devices

4 ( Similar to the actual

locations)

Trials

20

Size of grid

In Fig. 1 and 2, we can see that the proposedAIESRSSalgorithm performs better than the existing standard
algorithm in term localization error.
Localization error is computed in terms of RMSE as:
(9)
Where the number of nodes,
is the actual location
of the sensors or nodes,
is the estimated location of the
nodes.Average improvement in the minimization of
localization error can be seen in figure 1.

The adaptive information matrix is estimated as

(6)

The diagonal elements

of reduce to a single sum
over
since there are card
term in (1) that depend
. The off-diagonal elements when
as
on
(7)
The off-diagonal elements when

as
(8)

Where is an indicator functionn

or else it is 0.
III.

, where

SIMULATION RESULT AND ANNALYSIS

The system environment used is windows 7 enterprises 64bit operating system with 4GB of RAM. We have used
MATLAB tool. We have conducted simulation study on
following parameter for localization andpath gain and
compared our proposed AIES-RSS with standard localization

Fig. 1.Localization performance

In this case of localization error, siimulation study shows that

average error of localization of the existing model is 0.307and
by using proposed model this error is reduced to the 0.2891.

Third International Conference on Devices, Circuits and Systems (ICDCS'16)

Below given Fig. 2. shows the performance of the proposed
model in terms of the path gain. In this simulation we have
studied the path gain based on the time delay for varied number
of devices. The proposed model performs better and achieves
the better path gain when compared to the existing standard
algorithm.Simulation study shows that average path gain of the
existing model is 0.146769 and by using proposed model this
gain is is increased to the 0.046994283 which shows the
improvement of 32.019%.

[3]

[4]
[5]

[6]

[7]

[8]

[9]

[10]
[11]

[12]

[13]

Fig. 2.Gain Performance

IV.

CONCLUSION

Wireless sensor network is a network that enables

correspondence between various devices associated through a
infrastructure protocol. Finding the position or location of
sensor node (Localization) is an important factor in sensor
network for proving efficient service to end user. The existing
localization is not cost effective and efficient and it suffers
from localization optimization overhead.The existing technique
proposed so for suffers in estimating the likelihood of
localization error. To cater this in this work the author proposes
a RSS (Received signal strength) based localization technique
and also proposes an adaptive information estimation to reduce
or approximate the localization error in wireless sensor
network. The simulation results show the effectiveness of our
proposed localization model over existing protocol.

REFERENCES
[1]

[2]

L. Xu, and X. Wang, Compressed sensing signal and data acquisitionin

wireless sensor networks and Internet of Things, IEEE Trans. Ind.
Informat., vol. 9, no. 4, pp. 21772186, 2013.
D. Guinard, V. Trifa, S. Karnouskos, P. Spiess, and D. Savio,
Interactingwith the SOA-based Internet of Things: Discovery, query,

[14]

[15]

[16]
[17]

[18]

[19]

[20]

173

selection, and on-demand provisioning ofWeb services, IEEE Trans.

Services Comput., vol. 3, no. 3, pp. 223235, 2010.
D. Chiang, C. Lin, and M. Chen, The adaptive approach for storage
assignment by mining data of warehouse management system for
distribution centres, Enterprise Inf. Syst., vol. 5, no. 2, pp. 219234,
2011.
L. Duan and L. D. Xu, Business intelligence for enterprise systems: A
survey, IEEE Trans. Ind. Informat., vol. 8, no. 3, pp. 679687, 2012.
L. D. Xu,W. Viriyasitavat, P. Ruchikachorn, and A. Martin, Using
propositional logic for requirements verification of service workflow,
IEEE Trans. Ind. Informat., vol. 8, no. 3, pp. 639646, 2012.
E. Xu, M. Wermus, and D. Bauman, Development of an integrated
medical supply chain information system, Enterprise Inf. Syst., vol. 5,
no. 3, pp. 385399, 2011.
L. Duan, W. Street, and E. Xu, Healthcare information systems: Data
mining methods in the creation of a clinical recommender system,
Enterprise Inf. Syst., vol. 5, no. 2, pp. 169181, 2011.
P. Biswas, K.-C. Toh, and Y. Ye, A distributed SDP approach for
largescalenoisy anchor-free graph realization with applications to
molecular conformation, SIAM J. Sci. Comput., vol. 30, no. 3, pp.
12511277, Mar. 2008.
A. Boukerche, H. A. B. F. Oliveira, E. F. Nakamura, and A. A. F.
Loureiro,Localization systems for wireless sensor networks, IEEE
Wireless Commun., vol. 14, no. 6, pp. 612, 2007.
L. D. Xu, Enterprise systems: State-of-the-art and future trends, IEEE
Trans. Ind. Informat., vol. 7, no. 4, pp. 630640, 2011.
X. Wang and S. Li, Scalable routing modeling for wireless ad hoc
networks by using polychromatic sets, IEEE Syst. J., vol. 7, no. 1, pp.
5058, 2013.
O.-H. Kwon and H.-J. Song, Localization through map stitching in
wirelesssensor networks, IEEE Trans. Parallel Distrib. Syst., vol. 19,
no. 1, pp. 93105, Jan. 2008.
M. Hamdi, N. Boudriga, and M. S. Obaidat, Bandwidth-effective
designof a satellite-based hybrid wireless sensor network for mobile
target detection and tracking, IEEE Syst. J., vol. 2, no. 1, pp. 7482,
2008.
S. Lee, E. Kim, C. Kim, and K. Kim, Localization with a mobile
beaconbased on geometric constraints in wireless sensor networks,
IEEE Trans. Wireless Commun., vol. 8, no. 12, pp. 58015805, Dec.
2009.
S. Misra, S. K. Dhurandher, M. S. Obaidat, N. Nangia, N. Bhardway, P.
Goyal, and S. Aggarwal, Node stability-based location updating in
mobile Ad-Hoc networks, IEEE Syst. J., vol. 2, no. 2, pp. 237
247,2008.
T. S. Rappaport, Wireless Communications: Principles and
Practice.Englewood Cliffs, NJ: Prentice-Hall, 1996.
Weile Zhang; Qinye Yin; Hongyang Chen; Feifei Gao; Ansari, N.,
"Distributed Angle Estimation for Localization in Wireless Sensor
Networks," in Wireless Communications, IEEE Transactions on , vol.12,
no.2, pp.527-537, 2013
Malajner, M.; Gleich, D.; Planinsic, P., "Angle of Arrival Measurement
Using Multiple Static Monopole Antennas," in Sensors Journal, IEEE ,
vol.15, no.6, pp.3328-3337, 2015.
Quanrui Wei; Dexing Zhong; Jiuqiang Han, "Improved localisation
method based on multi-hop distance unbiased estimation," in
Communications, IET , vol.8, no.16, pp.2797-2804, 11 6 2014.
Maddumabandara, A.; Leung, H.; Minxiang Liu, "Experimental
Evaluation of Indoor Localization Using Wireless Sensor Networks," in
Sensors Journal, IEEE , vol.15, no.9, pp.5228-5237, Sept. 2015.