International Journal of Applied Engineering Research ISSN 0973-4562 Volume 13, Number 21 (2018) pp.
14964-14968
© Research India Publications. http://www.ripublication.com
A Study on Performance Analysis of Distance Estimation RSSI in Wireless
Sensor Networks
S.Satheesh1, Dr.V.Vinoba2
1
Assistant professor, T.J.S. Engineering College, Chennai-601206, Tamil Nadu, India.
2
Assistant professor, K.N. Government Arts College for Women (Autonomous), Thanjavur, Tamil Nadu, India.
Abstract information on the coordinates of a target placed on the
network; in fact, the fields of application are huge. Positioning
Research has shown that the awareness of positions of
devices are now part of the daily life of the population.
wireless sensor nodes is a desirable feature for many
applications in Wireless Sensor Networks (WSN). The
performance of analysis distance estimation in WSNs is the
LOCALIZATION IN WIRELESS SENSOR NETWORKS
association bordered by the Received Signal Strength
Indication (RSSI) values and distance. The RSSI of nets make Performance Metrics.
available a practical way of estimating the distance between
nodes because the use of it does not require any additional Multiple metrics can be used to measure the performance of a
hardware but simply a radio transceiver compared to other localization technique. It is not enough to observe accuracy
only. Referring to the literature and considering the results of
range based models. In this paper, Performance analysis of the
our research we provide the following performance measures:
RSSI model that estimates the distance between sensor nodes
accuracy, coverage, complexity, scalability, robustness, and
in WSNs is presented. It is shown that the results of this
evaluation can contribute towards obtaining accurate locations cost. They are mainly connected with economical or technical
of the wireless sensor. constraints such as hardware cost, low battery power, and
limited computation capabilities.
Keywords: wireless sensor network, RSSI model, distance
estimation, node.
Localization Accuracy.
Accuracy is the most important requirement of location
INTRODUCTION systems. Usually, the mean error between the estimated and
the true location of the co-anchor nodes in the network is
In sensor networks, 802.15.4 WPAN the most common
adopted as the performance metric. It is defined as follows:
technique used to calculate the distance between two nodes is
the RSSI (Received Signal Strength Indicator) technique
because it has the advantage of not requiring additional 2
hardware and synchronization on nodes. Some studies shows 1 ‖𝑅𝑥𝑖 −𝑅𝑥𝑗 ‖
LA = ∑𝑁 100%
that the RSSI index is fairly unreliable and often produces 𝑁 𝑖,𝑗=1 𝑅𝑖2
significant errors about the location of the nodes in the
network. In this paper, an approach to the problem of nodes
localization in an outdoor environment is proposed. In order Where N denotes the number of nodes in a network whose
to obtain more accurate distance estimation, a scenario location is estimated, LA denotes a localization error, 𝑅𝑥𝑗 the
dependent ranging technique has been adopted. The goodness true position of the node i in the network, 𝑅𝑥𝑗 is the estimated
of the ranging model is estimated through a comparison with
location of the node i (solution of the location system) and 𝑅𝑖
the classic model based on the path-loss long-distance; then
is the radio transmission range of the node i. The localization
two localization techniques such as Triangulation and Roc
error LE is expressed as a percentage error. It is normalized
RSSI are used in order to test the improvement obtained for
with respect to the radio range to allow comparison of results
the estimated positions of nodes within the network.
obtained for different sizes and ranges of networks. Usually,
Network evolution has experienced continuous and rapid centralized location systems give more accurate position
technological development in recent years. The concept of estimates than distributed ones. The Distributed
networking as a simple connection between terminals has implementation may involve a loss of information due to an
evolved, becoming increasingly sophisticated and detailed. incomplete network map and parallel computations.
Wireless Sensor Networks can observe and extract
It is obvious that the higher accuracy, the better the system.
information relating to the environment in which they are
However, there is often a trade-off between position
placed. Many studies about energy optimization, routing
estimation accuracy and other characteristics. Therefore a
information, data representation and more, be situated of
compromise between the required accuracy and other
current interest, and the location is a particularly interesting
characteristics is needed.
argument today. Attention to this issue is justified by the
increasing demand for more applications dealing with
14964
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 13, Number 21 (2018) pp. 14964-14968
© Research India Publications. http://www.ripublication.com
dense network, wireless signal channels may become
congested and more complex communication infrastructure
may be required. The location system can locate the nodes in
2- D or 3-D space; some of them can support both 2-D and 3-
D spaces. Consolidated systems usually aggregate all
measurements and input data at a central unit to carry out
processing. By contrast, distributed implementation of
localization improves scalability.
Definition 1
Interval data: For given 𝐼𝑑 ,𝐼𝑅 ∈ 𝑅, and𝐼𝑅 ≥ 𝐼𝑑 , we call the
set I = [𝐼𝑑 ,𝐼𝑅 ]={u/ 𝐼𝑑 ≤ 𝑢 ≤ 𝐼𝑅 }interval data, where 𝐼𝑑 is the
lower bound of the interval data, and𝐼𝑅 is the upper bound. If
𝐼𝑑 =𝐼𝑅 which means the upper and lower bounds are equal, the
interval data becomes exact data.
Definition 2
Figure 1. Localization scheme
Midpoint and radius of interval data: For a given interval
data I = [𝐼𝑑 ,𝐼𝑅 ], let𝑟𝐷 = [𝐼𝑑 − 𝐼𝑅 ] thus, we have
Coverage. 𝐼𝐷 = 𝐷𝑑 − 𝑅𝐷 , 𝐼𝐷 = 𝐷𝑑 + 𝑅𝐷
The coverage of localization procedures is related to the We define 𝐷𝑑 and 𝑟𝐷 (𝑟𝐷 ≥ 0) as the midpoint and radius,
deployment area, network density, hardware tools and respectively, of interval data I. Therefore, we can also express
resources of devices that form a network. Now and then in the interval data as follows:
effect in outsized, distributed sensor networks when nodes do
not have enough neighboring nodes, unevenly distributed
anchor nodes, or in the case of under the weather equipped 𝐼𝐷 = 𝐷𝑑 − 𝑅𝐷 , 𝐼𝐷 = 𝐷𝑑 + 𝑅𝐷
devices, problems with localization of the whole linkage may
occur. In such a situation the question is how much of the
network can be localized. In the case of poor results, the only Because we estimate RSSI-D according to the exact RSSI
option is to increase the number of anchor nodes in a network. values measured in the RSSI-D procedure, we propose our
third definition as the distance between the interval data and
the exact data.
Complexity.
(3) Distance between the interval data and the exact data:
The complexity of a location system can be attributed to
hardware, software, and operation factors. In general, range- For given interval data 𝐼𝑋 = 𝐷𝑋 − 𝑅𝑋 , 𝐷𝑋 + 𝑅𝑋 , y=y,
based methods are much complex than range-free techniques where𝑅𝑋 , 𝐷𝑋 ∈ 𝑅. The distance relationship between the two
and involve hardware complexity. Software complexity data sets is illustrated in Figure 1. When they are separate
depends on the computing complexity of the positioning from each other, as shown in Figure2 (a), the minimum
algorithm. In centralized location systems, a central unit distance is [𝐷𝑋 − 𝑌]-𝑅𝑋 and the maximum distance is [𝐷𝑋 −
calculates the estimated locations due to its powerful 𝑌]+𝑅𝑋 when they are joined, as shown in Figure 2(b), the
processing capability, and sufficient power supply and minimum distance is 0, and the maximum distance is [𝐷𝑋 −
memory. If calculations are carried out on the sensor node, the 𝑌]-𝑅𝑋 = 2𝑟𝑥 when the interval data contains the exact data, as
effects of complexity could be evident. The Most procedures shown in Figure 2(c), the minimum distance is 0, and the
that form a sensor network lack strong processing power, maximum distance is [𝐷𝑋 − 𝑌]+𝑅𝑋 = 2𝑟𝑥 . Therefore, we can
memory and power source, so techniques with low complexity calculate the maximum distance 𝐷𝑚𝑎𝑥 between X and Y, the
are often preferred. minimum distance 𝐷𝑚𝑖𝑛 and the distance d between the
interval data and the exact data as follows:
Scalability. 𝐷𝑚𝑖𝑛 = max(0, |[𝐷𝑋 − 𝑌]| − 𝑅𝑋 )
The scalability of a location system ensures suitable 𝐷𝑚𝑎𝑥 = |[𝐷𝑋 − 𝑌]| + 𝑅𝑋
estimation of localization when the network or deployment
D= [𝐷𝑚𝑖𝑛 𝐷𝑚𝑎𝑥 ]
area gets larger. A location system should scale on the
network size (number of nodes) and density, the size of a As indicated by Equation, the distance between the interval
deployment area and dimensional space. In the case of range- data and the exact data remains as interval data, which can
based techniques, the location performance degrades when the comprehensively represent different distance values.
distance between the transmitter and receiver increases. The
14965
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 13, Number 21 (2018) pp. 14964-14968
© Research India Publications. http://www.ripublication.com
On-line distance estimation: During the RSSI-D estimation
technique, the RSSI charge is restrained by a WSNs (e.g.,
DD2530 WSN node), and we can estimation the communiqué
detachment using indeterminate data gathering.
In the on-line distance estimation component, considering
poles apart stages of improbability in RSSI values, we accept
RSSI-D estimation ways and means using both hard and easy-
going uncertain data gathering methods to increase the
approximation accuracy.
The hand-outs of this manuscript are as follows:
We propose DEUDC, a RSSI-based communication
estimation method, which uses a mapping strategy and an
uncertain data clustering method. Unlike sample-based
mapping in RADAR and ARIADNE systems, we resort
to distribution-based mapping to overcome the
uncertainty in RSSI readings.
To address the uncertainty in RSSI values, we adopt
interval data and statistical information to represent the
RSSI distribution characteristic of each distance. In
comparison to sample-based mapping, by exploiting
distribution-based statistics, our approach can potentially
obtain greater improvement in estimation accuracy and
Figure 2. Distance Relation between Interval Data efficiency.
and Exact Data We propose an RSSI-D estimation method in which
uncertain data soft and hard clustering algorithms are
implemented in order to obtain better estimation accuracy
DISTANCE ESTIMATION IN ONLINE AND OFF-LINE with respect to different levels of uncertainty in RSSI.
We have evaluated DEUDC using real data sets from
To improve distance estimation accurateness, we have representative wireless environment. Experimental results
wished-for a RSSI-D approximation technique using interval show that DEUDC out-performs state-of-art estimation
data gathering, called Distance Estimation using methods.
Indeterminate Data Gathering (DEUDC). As given away in
Figure 3, the background of DEUDC is encompassed by an
off-line environment measurement component and an online
distance estimation module.
Off-line environment measurement: We first complete RSSI
illustration measurements at poles apart to improve distance
estimation accurateness, we have wished-for an RSSI-D
approximation technique using interval data gathering, called
Distance Estimation using Indeterminate Data Gathering
(DEUDC). As given away in Figure 3, the background of
DEUDC is encompassed by an off-line environment
measurement component and an online distance estimation
module communiqué points in the wireless communiqué
surroundings. We then deference to the RSSI data intended
for arithmetical computation and model the RSSI distribution
distinguishing in terms of RSSI uncertainties. We can obtain
an RSSI-D charting grounded on this technique.
Figure 4 .Impact of Correlation Factor on the RSSI-D
Estimation Method
Figure 3. The framework of DEUDC.
14966
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 13, Number 21 (2018) pp. 14964-14968
© Research India Publications. http://www.ripublication.com
Parameter Corridor Hall Air REFERENCES
Node DD2530 DD2530 DD2530 [1] I.F. Akyildiz,W.Su, Sankarasubramaniam, and Cayirci,
Temperature 22 0 c0
22.5 0 c0 0
20 0 c “Wireless sensor network: a survey, “Computer
Networks (Elsevier), Vol.38, No.4, 393-422, 2002.
RSSI-D estimation 3mx3m 3mx3m 3mx3m
RSSI-D estimation points 50 50 50 [2] Guoqiang Mao,Baris Fidan,Brian D.O.
Anderon,Wireless sensor network localization
Height of node 0.1 m techniques,Computer Networks, Vol.51, 2529-2553,
Table 1. Experimental temperature conditions and parameters 2007.
[3] Patwari N,Aah J N,Kyperountas Setallocating the
nodes cooperative localization in wireless sensor
networks,IEEE Signal Processing Magazine, Vol.22,
No.4, 54-69, 2005.
[4] N. Patwari,A. Hero,M. Perkins, N,et al.Relative
location estimation in wireless sensor networks,IEEE
Transactions on Signal Processing, Vol.51, No.8,2137-
2148, 2003.
[5] LI Yaoyi,HE Xiaoxing ,LIU Shouyin,Wireless
Localizaion Algorithm Based on Path Loss Model
Parameter Estimated in Real-time [J],CHINESE
JOURNAL OF SENSOR AND CATUATORS, Vol.23,
No.9, 1328-1333, 2010.
[6] GAO Guosheng, CHEN Junjie,LI Gang. Beacon Node
Self-Tuning Localization Algorithm Based on RSSI,
Figure 5. Implementation of Distributed DEUDC RSSI-D Measurement & Control Technology, Vol28, No.8, 93-
Estimation 97, 2009.
[7] L iD,Wong KD, Hu Y H, et al. Detection,
classification, and tracking of targets . IEEE Signal
Processing Magazine,2002, 19 (2) : 17 - 29.
[8] Chen Weike,Li Wenfeng,Shou Heng,etc.,Weighted
Centroid Localization Algorithm Based on RSSI for
Wireless Sensor Networks,Journal of Wuhan
University of Technology(Transportation
Science&Engineer), Vol.30, No.2, 265-268, 2006.
[9] Fugen Su, Weizheng Ren,Hongli Jin.Localization
algorithm based on difference estimation for wireless
sensor networks. Communication Software and
Networks,2009,ICCSN’09, International Conference
on,27-28 Feb. 2009 Page(s):499 – 503.
[10] Kumar S, Lobiyal D K，“Novel DV-Hop localization
algorithm for wireless sensor networks”
Figure 6. Changes in distance estimation with coverage factor Telecommunication Systems, 2016: 1-16.
in three environments.
[11] Al Alawi R, “RSSI based location estimation in
wireless sensors networks” 2011 17th IEEE
International Conference on Networks. IEEE, pp: 118-
CONCLUSION
122, 2011.
In this better-quality technique of RSSI capacity has been
[12] Lim C B, Kang S H, Cho H H, “An Enhanced Indoor
contemporary. Experimental measurement and simulation
Localization algorithm Based on IEEE 802.11 WLAN
results show that the computation time has been increased, but
using RSSI and multiple parameters”2010 Fifth
measurement accuracy can be improved greatly. The proposed
International Conference on Systems and Networks
method can reduce the error of RSSI measurement, which can
improve the localization accuracy. This measurement method Communications. IEEE, pp.238-242,2010.
is an option in wireless sensor node localization. [13] Li J, Zhong X, Lu I T, “Three-dimensional node
localization algorithm for WSN based on differential
14967
International Journal of Applied Engineering Research ISSN 0973-4562 Volume 13, Number 21 (2018) pp. 14964-14968
© Research India Publications. http://www.ripublication.com
RSS irregular transmission model”，Journal of
Communications, vol.9,no.5, pp.391-397,2014.
[14] Srbinovska M, Dimcev V, Gavrovski C, et al
“Localization Techniques in Wireless Sensor Networks
using Measurement of Received Signal Strength
Indicator”, FACULTY OF ELECTRICAL
ENGINEERING UNIVERSITY OF BANJA LUKA,
vol.15,no.1,pp.67-71,2011.
[15] Heurtefeux K, Valois F. “Is RSSI a good choice for
localization in wireless sensor network”, IEEE 26th
International Conference on Advanced Information
Networking and Applications. IEEE, pp.732-739, 2012.
[16] Cao Z, “Experimental exploration of RSSI model for
the vehicle intelligent position system”, Journal of
Industrial Engineering and Management, vol.8, no.1,
pp.51-1, 2015
14968
