Sensors and Actuators A 271 (2018) 283–289

Contents lists available at ScienceDirect

Sensors and Actuators A: Physical

journal homepage: www.elsevier.com/locate/sna

Localization improvement in wireless sensor networks using a new

statistical channel model
Amir Karimi Alavijeh a,∗ , Mohammad Hossein Ramezani b , Ali Karimi Alavijeh c
a
Faculty of Electrical Engineering and Computer Science, University of S. Beheshti, Tehran, Iran
b
Mads Clausen Institute, Faculty of Engineering, University of Southern Denmark, Sønderborg, Denmark
c
Faculty of Electrical Engineering, Azad University of Najafabad, Isfahan, Iran

a r t i c l e i n f o a b s t r a c t

Article history: In this paper, a statistical channel model is proposed based on the second moment of Received Signal
Received 12 July 2017 Strength Indicator (RSSI) in an outdoor communication channel. The medium under study is a grass field
Received in revised form where the RSSI data are collected in different distances and orientations using a set of in-house built
22 December 2017
transmitter-receiver sensors. The validity of the constructed sensors is confirmed since the first moment
Accepted 8 January 2018
Available online 31 January 2018
of RSSI data follows the well-known Friis model. The proposed model presents an additional relationship
between the variance of RSSI data and distance. To demonstrate the application of this statistical rela-
tionship, we have investigated the localization problem of a hidden node using extended Kalman filter
Keywords:
Localization (EKF). Compared to the conventional EKF in which the covariance matrix of measurement noise is fixed,
Channel modeling this matrix can be updated online using the proposed model. The experimental and simulation results of
Measurement noise two different scenarios, which are fixed hidden node and mobile hidden node, show that the proposed
Wireless sensor network model improves the accuracy of RSSI localization from 10 to 22 percent in different situations.
Extended Kalman filters © 2018 Elsevier B.V. All rights reserved.

1. Introduction
Localization of wireless sensors has received significant atten-
tion for the past few decades and has been exploited in many
applications such as environmental monitoring [1], habitat mon-
itoring [2], mobile robot tracking [3], wireless video capsule
endoscopy [4], and military applications [5]. It has always been
essential for the humankind to gather information from the envi-
ronment in order to make a proper decision about what has
happened or is going to happen. However, data gathering in many
places is a dangerous, difficult, or time consuming job, and wire-
less sensor networks (WSNs) are one of the best substitutions in
such conditions. Thus the wireless sensors of a WSN are sometimes
inaccessible or hidden from sight.
So far different localization algorithms have been proposed and
applied considering energy efficiency, hardware limitations, sim-
plicity, performance, accuracy and so on [6,7]. Ref. [8] presented
three basic evaluation metrics to compare the performance of local-
ization techniques which are localization accuracy, computation
and communication costs, and network and anchors density. Nev-
ertheless, it cannot be said what algorithm is the best because the
localization algorithms are severely affected by the application sce-
nario and each one can be the best in a certain application scenario.
One main category of localization algorithms is called range-
based algorithms in which the WSN consists of some anchors,
which are wireless sensors with known coordinates, and a hidden
node, which its coordinate is going to be localized. Range-based
algorithms generally include two phases which are ranging and
location computation [9]. Time of Arrival (TOA), Time Difference of
Arrival (TDOA), Angle of Arrival (AOA), and Received Signal Strength
Indicator (RSSI) are some of the measurement techniques that are
used in the first phase [10]. However, RSSI is one of the cheapest
measurement techniques since there is no need to add any extra
devices [11]. In this method, which has been widely used, the RSSI
signal is measured based on the received signal strength as follows
[12]:
 Received Power in Watt 
RSSI (dBm) = 10 log (1)
0.001 Watt

RSSI is based on the fact that the received signal strength at the
∗ Corresponding author.
receiver is affected by the attenuation of the medium which is in
E-mail addresses: am.ka1981@gmail.com (A. Karimi Alavijeh),
turn a function of distance and the medium characteristics [13].
ramezani@mci.sdu.dk (M.H. Ramezani), ali.karimi6586@gmail.com However, this method suffers greatly from noise. Channel modeling
(A. Karimi Alavijeh). is an approach to study the behavior of measurement noise. Having

https://doi.org/10.1016/j.sna.2018.01.015
0924-4247/© 2018 Elsevier B.V. All rights reserved.
284 A. Karimi Alavijeh et al. / Sensors and Actuators A 271 (2018) 283–289

such a model, one can incorporate the statistical characteristics of Table 1

Specifications of XBee-PRO S1 module [22].
noise into estimation of distance [14,15].
Concerning the second phase of localization, geometrical algo- Outdoor/RF line-of-sight range 1.6 km
rithms such as lateration are the primary localization algorithms. Transmit power 60 mW
Serial data interface 3.3V CMOS UART
In this case, regardless of the direction of the wireless sensors,
Frequency band 2.4 GHz
the coordinates of at least n + 1 anchors are needed for localiza- Supply voltage 2.8–3.4VDC
tion on a n-dimension surface [16]. Lateration can be divided into Transmit current 215 mA @ 3.3VDC
two groups: trilateration and multilateration. In trilatertion, con- Receive current 55 mA @ 3.3VDC
sidering a two-dimensional surface, the exact distance between a
hidden node and three anchors is needed to determine the unique
position of the hidden node. It should be noted that by assuming a
2D surface, all of the sensors such as hidden node and anchors are in
the same altitude. However, the exact distances between anchors
and hidden node cannot be acquired in the natural presence of mea-
surement uncertainty. On the other hand, multilateration is used to
reduce the impact of measurement uncertainty and obtain better
position estimation by employing more anchors; but nonethe-
less, there is still a noticeable error between the estimated and
actual position. Thus, some other algorithms have been proposed
to decrease such error.
In order to improve the localization results of lateration, this
algorithm was combined with some other algorithms. For instance,
[17] implemented three different methods, multilateration, piece-
wise maximum likelihood centroid, and EKF, and concluded that
EKF would result in a better estimation. There are a variety of other
references such as [18,19] that applied EKF for localization. The EKF
in Ref. [19] used a method in which EKF was employed to filter the
measured RSSI values and converted it to distance.
The contribution of the current study is to propose a model
Fig. 1. Designed and developed wireless sensors.
expressing the relationship between the variance of RSSI and dis-
tance. Most of the related studies in the literature, such as [19,20],
which have used RSSI values for distance estimation, only take into
consideration the first moment of RSSI. However, as it is also shown loss model and the proposed channel model, are explained and
in [21], the second moment (variance) of RSSI can have some useful their parameters are identified. The mathematical description of
information regarding the distance. In other words, longer distance the system, which is presented in Sections 4 and 5, describes the
measurements have higher variances as a result of fading errors. experiments and simulation results. Finally, Section 6 is allocated
The property of proposed method is that it enables us to tune the to the conclusion.
covariance matrix of EKF adaptively corresponding to distance. As
a result, the closer anchors to the hidden node will have a higher
2. Experimental setup
weight in localization.
In order to validate the proposed model and compare the results
In order to gather experimental RSSI, a set of in-house built wire-
with previous algorithms, several steps have been accomplished in
less sensors were developed. Each of the sensors features a board
this study. First, the RSSI signal is gathered in determined distances
that consists of:
on a grass football pitch by utilizing two in-house built wireless sen-
sors. Then the validity of the hardware and measuring method is
confirmed by comparing the first moment of the RSSI with the Friis • a XBee-PRO 802.15.4 module (Series 1) with the main specifica-
channel model reported in the literature [14,15]. Second, a logarith- tions listed in Table 1
mic relationship between the variance of RSSI data and the distance • a 2.4 GHz omnidirectional 1.2 dBi antenna
is suggested, then the parameters of this model are estimated using • six LED indicators
the measured data. Finally, the proposed model is used to automat- • an Atmega32 micro-controller as a CPU in the transmitter
ically tune the covariance matrix of EKF, which here after is called • a connection between the receiver and the laptop by using an
varying covariance EKF (VCEKF), and the results are compared with RS232-to-USB converter cable
the constant covariance EKF (CCEKF). • a 6.2 VDC battery in the transmitter which feeds the module (by
In the experiments/simulations, we will consider two scenarios using a 3.3-V regulator) and the micro-controller (by using a 5-V
to have more general investigation; (1) fixed hidden node and (2) regulator).
moving hidden node. For the first scenario we have real data, since
we have measured RSSI in many different distances and angles. Since the receiver is connected to the laptop, it is fed directly
However the second scenario will be performed based on the sim- by another USB port and does not need any extra power supply.
ulation. In order to confirm the reliability of results of the second Fig. 1 shows the transmitter sensor in which the main elements are
scenario, the first scenario is also performed based on the simula- pointed.
tion data besides experimental data. The results of applying CCEKF
and VCEKF in both scenarios confirm the impact of the proposed
channel model in improving the precision of the localization. 3. Wireless communication channel modeling
The rest of the paper is organized as follows. Section 2 describes
the in-house built wireless sensors and their components. Section This section deals with two different models of the wireless
3 describes the channel modeling in which both models, the path communication channel.
 A. Karimi Alavijeh et al. / Sensors and Actuators A 271 (2018) 283–289 285

Fig. 3. Path loss model.

Fig. 2. Different positions of transmitter around the receiver where the data were
gathered.

3.1. Path loss model

As transmitted signal travels through transmission channel,

which is the space between transmitter and receiver, the strength
of signal is attenuated. Despite other impacts of the channel on
the wireless signal such as shadowing and multipath, the distribu-
tion of the propagated signal can be approximated as log-normal.
This simple model, which is widely used in outdoor and indoor
applications [14,15,23], is represented by the following equation:
d
PL (d) = P̄L (d0 ) − 10n log + (2)
d0

where PL (d) is the RSSI at the receiver, P̄L (d0 ) is the average RSSI
strength at the reference distance d0 , n is the path loss exponent, d
is the distance between transmitter and receiver, and  is a logarith-
mic Gaussian noise that models fading. Having wide measurements
in different distances, the parameters P̄L (d0 ) and n can be obtained
by minimizing the mean square error (MSE) between the model
output and the empirical measurements [14]. The transmitter was
placed around the receiver at angles 0◦ , 90◦ , 180◦ , and 270◦ to deal
with the heterogeneity of the antenna, and 11 determined distances
10 m, 20 m, up to 110 m. The RSSI was then gathered each time for
100 s with sample time 0.1 s (1000 samples). Fig. 2 shows the dif-
ferent positions of transmitter around the receiver where the data
were gathered.
A curve with the form of (2) was then fitted on them, and the
parameters of the model were calculated as below:

P¯L (d0 ) = −41.5194 dBm, n = 3.3957. (3)

The obtained value of n is in the reported feasible interval [14],

which confirms the validity of the sensors. Fig. 3 shows the average
of RSSI at each distance for different angles.

3.2. Variance of RSSI versus distance

As a sample of measured data, Fig. 4 shows the RSSI data at 70 m

(90◦ ) and its histogram and covariance matrix. To investigate the
assumptions regarding  in (2), we have fitted a Gaussian distribu-
tion on the data in Fig. 4b. As it can be seen from the fitted curve Fig. 4. (a) Typical measured RSSI, (b) its histogram, and (c) its auto-covariance.

and the pseudo impulse behavior of the measured auto-covariance,

286 A. Karimi Alavijeh et al. / Sensors and Actuators A 271 (2018) 283–289

Fig. 6. The hidden node and the anchors.

constant velocity, constant acceleration and colored noise accel-

eration [17,24,25]. In this study, by assuming a time varying
Fig. 5. The path loss second moment model. acceleration without sudden changes, the colored noise acceler-
ation approach is considered. In this approach, the state vector
the assumptions of having a white Gaussian noise are acceptable is:
to some extent.  T
x̂[0] = mx , vx , ax , my , vy , ay (7)
The other important point, which can be understood from the
gathered data and can be noticed in Fig. 5, is that as the distance where vx and ax are the speed and the acceleration of m along x
between transmitter and receiver increases, the variance of RSSI (horizontal) axis, respectively, and so are vy and ay but along y (ver-
goes up. This result is intuitively logical because of the presence tical) axis. The motion equations of the system can be formed using
of the attenuation factors such as reflection. In order to estimate classical mechanics:
a relationship between the variance of RSSI and distance, the vari-
ance of the data (containing 1000 samples) in each position of the ṁx = vx , v̇x = ax
(8)
transmitter is calculated and plotted in Fig. 5. The variance of the ṁy = vy , v̇y = ay .
whole data in each distance (containing 4000 samples) is also high-
lighted in this figure. Considering the logarithmic axes of the figure, Considering the acceleration to be a first order colored noise, two
a logarithmic relation between the variance of RSSI and the distance other state equations can be written as follows:
is imaginable as below: −1 1
ȧx = ax + x
VRSSI = a + b log(d) (4) x x
(9)
−1 1
in which VRSSI is the variance of RSSI at the distance d, while a and ȧy = ay + y
y y
b are the constant parameters of the model. The values of these
parameters are obtained by curve fitting as follows: where  x and  y are the time constants of the first order filters, and
 x and  y are independent white Gaussian noise processes along
a = −2.7959, b = 1.6463. (5)
the x and y axes, respectively:
The outlier values of variance at 10 m are caused by signal strength
x ∼N(0, Qx )
saturation for the short distances between transmitter and receiver. (10)
Compared with the path loss model in Fig. 3, the new model of the y ∼N(0, Qy )
channel can be called: “path loss second moment model”.
in which Qx and Qy are covariance matrices. The whole equations
in matrix form are as follows:
4. Mathematical description of the system
⎡ ṁx ⎤ ⎡ ⎤ ⎡ mx ⎤ ⎡ ⎤
0 1 0 0 0 0 0 0
In this section, we consider a WSN which consists of q fixed ⎢ v̇x ⎥ ⎢ 0 0 1 0 0 0 ⎥ ⎢ vx ⎥ ⎢ 0 0 ⎥
⎢ ⎥ ⎢ −1 ⎥⎢ ⎥ ⎢ 1 ⎥ 
anchors, a1 , a2 , a3 , . . ., and aq with known coordinates, and a fixed ⎢ ȧx ⎥ ⎢ 0 0 0 0 0 ⎥ ⎢ ax ⎥ ⎢  0 ⎥ x
or mobile hidden node, m. The localization process is defined as ⎢ ⎥=⎢ x ⎥⎢ ⎥ + ⎢ x ⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ (11)
estimation of the instantaneous position of m in Cartesian coordi- ⎢ ṁy ⎥ ⎢ 0 0 0 0 1 0 ⎥ ⎢ my ⎥ ⎢ 0 0 ⎥ y
⎢ ⎥ ⎢0 ⎥⎢ ⎥ ⎢ 0 ⎥
nate, i.e. mx and my , using the measured RSSI values between the ⎣ v̇y ⎦ ⎣ 0 0 0 0 1 ⎦ ⎣ vy ⎦ ⎣ 0 ⎦
hidden node and the anchors. In this paper, we consider the 2D −1 1
0 0 0 0 0 0
ȧy y ay y
problem, nevertheless the method is readily extendable to the 3D
localization problem [4]. For the fixed hidden node, the velocity is zero, so the equations of
For the sake of simplicity, first let q = 3, so the known coordinates the system are simplified as below:
of a1 , a2 , and a3 are (x1 , y1 ), (x2 , y2 ), and (x3 , y3 ), respectively (Fig. 6). ⎡ ⎤ ⎡ ⎤
Let di be the Euclidean distance between m and ai , so: ṁx ⎡ ⎤ mx ⎡ ⎤
0 1 0 0 0 0  
 ⎢ ⎥ ⎢ ⎥
 2 ⎢ v̇x ⎥ ⎢ 0 0 0 0 ⎥ ⎢ vx ⎥ ⎢ 1 0 ⎥ x
di = (mx − xi )2 + my − yi i = 1, 2, 3. ⎢ ⎥=⎢ ⎥⎢ ⎥ ⎢ ⎥
⎢ ṁ ⎥ ⎣ 0 0 0 1 ⎦ ⎢ m ⎥ + ⎣ 0 0 ⎦ 
, (6) (12)
⎣ y⎦ ⎣ y⎦ y
First, suppose that the hidden node is moving. There are dif- 0 0 0 0 0 1
ferent models to describe the motion of hidden node such as v̇y vy
 A. Karimi Alavijeh et al. / Sensors and Actuators A 271 (2018) 283–289 287

Discretizing the state equations (12) and (11), we have:

xk = Axk−1 + B (13)
k−1

where xk ∈ Rp and xk−1 ∈ Rp are state vectors at time steps k and

k − 1, respectively, and  ∈ Rl is the noise process that is sup-
k
posed to be zero mean white Gaussian with covariance matrix
Q k ∈ Rl×l . The measurement equation of the system is the rela-
tionship between RSSI and the distance between the anchors and
the hidden node. Considering z k being the measured RSSI and using
(2) and (3), the measurement equations will be obtained as follows:
 
2 2
(mx − xi ) + (my − yi )
zi = PL (di ) = −41.5194 − 33.957 log + i (14)
10

in which i = 1, 2, 3. Eq. (14) is a non-linear measurement equation

in the form of:
 
z k = h xk ,  k (15)

where z k ∈ Rq , is the vector of the measurements gathered by q

sensors at time step k,  k ∈ Rq is a zero mean white Gaussian
noise with covariance matrix Rk ∈ Rq×q that models measurement
uncertainty. In order to estimate (mx , my ), we will use EKF in this
study. In conventional localizations employing EKF, the measure-
ment covariance matrix is a tunable parameter of EKF which is
tuned off-line. In this paper, as an application example, we will Fig. 7. Localization of the fixed hidden node using experimental data. (a) The RMS of
show that the proposed model (4) can be used to automatically tune the localization error for the CCEKF and VCEKF algorithms vs. the number of anchors,
the covariance matrix and improve the localization estimation: (b) the percentage of the localization improvement.

R = diag (a + b log(di )) , i = 1, 2, 3, . . ., q.
5. Experimental and simulation results
This section provides the evaluation of the proposed chan-
nel model by defining two different scenarios; fixed hidden node
and mobile hidden node. In the first scenario, the hidden node is
fixed and we have experimental data (since we already have the
measured data at 44 different positions around the hidden node).
The second scenario evaluates the impact of using the proposed
model on the improvement of localization when the hidden node
is mobile. Since there are no measured data in this case, the eval-
uation is carried out by using simulation in which the behavior of
the measured noise is generalized to the whole area under study. In
order to validate the simulation results of the second scenario, the
first scenario is evaluated by simulated data as well as real data and
the results are compared. Achieving the similar results from exper-
iments and simulations in the first scenario confirms the validity of
the simulation. According to Fig. 2, a circular area of radius 110 m is
covered. In this area, the anchors are located at known positions but
the hidden node could be everywhere (in mobile case). However,
in our experiments, in the fixed hidden node scenario, the hidden
node is located at the origin, so the distance between hidden node
and anchors could vary from 10 to 110 and In the mobile hidden
node scenario, the distance could vary from 0 to 220 m. Basically we
have not fixed the distance between anchors and the hidden node.
Therefore, it should be noted that the errors reported in Fig. 7 are
not corresponding to specific distances.
There are real data for 44 different positions of the transmit-
ter which can be used for localization in the first scenario. Since
localization is possible by having at least 3 distances (if all of the
anchors are not along a line), we selected 1000 combinations of k
anchors (k = 3, 4, . . ., 10) randomly for the both experimental and
simulated data. All of the simulations in this paper were done in
MATLAB environment.
(16) The enhancement of localization is evaluated by the Root Mean
Square (RMS) of the position errors using (17):

 N
1
RMS loc =  m2act [k] − m2est [k] (17)
N
k=1
where N is the number of samples in each repeat of simulation
(N = 1000), and mact [k] and mest [k] are the actual and the estimated
positions of the hidden node, respectively. In order to investigate
the impact of using the proposed model, all estimations will be
performed by two algorithms; CCEKF and VCEKF. The covariance
matrix of measurement noise is constant in the CCEKF algorithm
while the VCEKF algorithm uses (16) to update this matrix on-line.
5.1. Scenario 1
The hidden node is fixed at the origin and its position is esti-
mated by both VCEKF and CCEKF with the same initiation:
x̂[0] = [−1, −1]T
P̂[0] = diag([1, 1]) (18)
Q = diag([0.003, 0.003]).
R is calculated by using (16) in VCEKF, while its elements are
constant in CCEKF but are calculated by using (16) at d = 55 m. The
results of 1000 times localization for each set of anchors are shown
in Fig. 7. As it can be seen, the localization error in VCEKF is less than
CCEKF while both of them are decreasing by increasing the num-
bers of anchors. The lower plot of Fig. 7 shows the improvement of
VCEKF vs. the number of anchors. Regarding this figure, the less the
number of anchors is, the more improvement can be seen.
In order to compare the results of localization by using the
experimental and the simulated RSSI, the hidden node at the ori-
gin is localized again with the same positions for anchors but using
simulated data. The simulated RSSI is generated by (14) with an
288 A. Karimi Alavijeh et al. / Sensors and Actuators A 271 (2018) 283–289

Fig. 8. Localization of the fixed hidden node using simulated data. (a) The RMS of the Fig. 9. Localization of the mobile hidden node using simulated data. (a) The RMS of
localization error for the CCEKF and VCEKF algorithms vs. the number of anchors, the localization error for the CCEKF and VCEKF algorithms vs. the number of anchors,
(b) the percentage of the localization improvement. (b) the percentage of the localization improvement.

additive Gaussian white noise whose covariance is determined by Fig. 9 depicts the RMS error of CCEKF and VCEKF and the improve-
(16). Fig. 8 shows the results of localization by employing simu- ment percentage of localization of VCEKF compared with CCEKF
lated RSSI. As it can be seen, the trends of reduction in localization for different numbers of anchors (3–10). The results show that the
error and localization improvement are almost the same as the localization improvement is also true for moving target.
experimental results. Although, as expected the absolute value of
localization error is significantly less than the first one since (14) 6. Conclusion
is an approximate model. In reality, there are many sources for the
noise term  in (2), such as orientation of the antennas, fading and In this paper, after gathering RSSI by employing wireless
reflection. Specially the orientation of antenna is really important sensors, two communication channel models based on the first
as it can be seen from the distribution of points in each vertical line moment and the second moment of RSSI data were extracted.
of Fig. 5. In order to generate noise  in simulations, we have con- The first model was exploited to validate the correct operation
sidered an ideal case which is a Gaussian noise with zero mean and of the developed sensors according to the obtained parameters of
variance given by (4). This causes a perfect match between model the channel. The second model showed that the variance of RSSI
noise and estimated noise, hence we have a very small error in sim- grows up as the distance increases. By proposing a mathematical
ulations. However, regarding the relative error between VCEKF and relationship for the dependency of RSSI variance and distance, we
CCEKF, it can still show the improvement which fairly matches the investigated a localization problem using EKF to present an appli-
experimental results. cation for the proposed model. It is shown that by on-line updating
of the EKF covariance matrix the localization can be improved com-
5.2. Scenario 2 pared to the conventional EKF. The results of the simulations can
be summarized as follows. First, localization of a fixed hidden node
The second scenario has the same condition as the first scenario using VCEKF resulted in about 22 percent improvement in compar-
except that it considers a moving hidden node. The hidden node ison with CCEKF in the case of having only three anchors. Second,
moves on a circle with the radius of 15 m with constant angular despite the number of anchors, the localization error of VCEKF is
speed. The governing equations of the system are as (11) and the always less than CCEKF algorithm. Third, localization improvement
initial values are assumed to be as below: in VCEKF is more obvious for lower number of anchors. Fourth and
finally, the improvement of the localization is also valid for mobile
x̂[0] = [17, 0.05, 0.05, 1, 0.1, 0.1]T hidden nodes.
 
P̂[0] = diag 1, 1, 1, 1, 1, 1 (19)
  References
Q = diag 0, 0, qxc , 0, 0, qyc
[1] H. Yang, Y. Qin, G. Feng, H. Ci, Online monitoring of geological CO2 storage and
where leakage based on wireless sensor networks, IEEE Sens. J. 13 (2) (2013)
556–562.
x = y = 4
[2] I. Stojmenovic, Handbook of Sensor Networks: Algorithms and Architectures,
1 (20) vol. 49, John Wiley & Sons, 2005.
qxc = qyc = × 3.5 × 10−3 [3] J. Zhang, S. Li, G. Lu, Q. Zhou, A new wireless sensor localization and pose
x 2 tracking system for an autonomous mobile robot, in: 2010 International
 A. Karimi Alavijeh et al. / Sensors and Actuators A 271 (2018) 283–289
Conference on Mechatronics and Automation (ICMA), IEEE, 2010, pp.
1971–1975.
[4] E.S. Nadimi, V. Tarokh, Bayesian source localization in networks with
heterogeneous transmission medium, Navigation 59 (3) (2012) 163–175.
[5] M.P. Ðurišić, Z. Tafa, G. Dimić, V. Milutinović, A survey of military applications
of wireless sensor networks, in: 2012 Mediterranean Conference on
Embedded Computing (MECO), IEEE, 2012, pp. 196–199.
[6] V. Garg, M. Jhamb, A review of wireless sensor network on localization
techniques, Int. J. Eng. Trends Technol. (IJETT) 4 (4) (2013).
[7] A. Mesmoudi, M. Feham, N. Labraoui, Wireless sensor networks localization
algorithms: a comprehensive survey, arXiv:1312.4082.
[8] J. Wang, R.K. Ghosh, S.K. Das, A survey on sensor localization, J. Control Theory
Appl. 8 (1) (2010) 2–11.
[9] J.-A. Jiang, X.-Y. Zheng, Y.-F. Chen, C.-H. Wang, P.-T. Chen, C.-L. Chuang, C.-P.
Chen, A distributed RSS-based localization using a dynamic circle expanding
mechanism, IEEE Sens. J. 13 (10) (2013) 3754–3766.
[10] B.-C. Seet, Q. Zhang, C.H. Foh, A.C. Fong, Hybrid RF mapping and Kalman
filtered spring relaxation for sensor network localization, IEEE Sens. J. 12 (5)
(2012) 1427–1435.
[11] H. Karl, A. Willig, Protocols and Architectures for Wireless Sensor Networks,
John Wiley & Sons, 2007.
[12] V. Apte, Y.A. Powar, et al., Improving the accuracy of wireless LAN based
location determination systems using Kalman filter and multiple observers,
in: Wireless Communications and Networking Conference, 2006. WCNC 2006.
IEEE, vol. 1, IEEE, 2006, pp. 463–468.
[13] T. Chuenurajit, D. Suroso, P. Cherntanomwong, Implementation of RSSI-based
3D indoor localization using wireless sensor networks based on ZigBee
standard, J. Inf. Sci. Technol. 3 (2) (2012).
[14] A. Goldsmith, Wireless Communications, Cambridge University Press, 2005.
[15] V. Erceg, L.J. Greenstein, S.Y. Tjandra, S.R. Parkoff, A. Gupta, B. Kulic, A.A. Julius,
R. Bianchi, An empirically based path loss model for wireless channels in
suburban environments, IEEE J. Sel. Areas Commun. 17 (7) (1999) 1205–1211.
[16] H. Lee, S. Lee, Y. Kim, H. Chong, Grouping multi-duolateration localization
using partial space information for indoor wireless sensor networks, IEEE
Trans. Consum. Electron. 55 (4) (2009).
[17] X. Li, Y. Zhang, K. Xu, G. Fan, H. Wu, Research of localization and tracking
algorithms based on wireless sensor network, J. Inf. Comput. Sci. 8 (4) (2011)
708–715.
[18] F. Gustafsson, F. Gunnarsson, N. Bergman, U. Forssell, J. Jansson, R. Karlsson,
P.-J. Nordlund, Particle filters for positioning, navigation, and tracking, IEEE
Trans. Signal Process. 50 (2) (2002) 425–437.
[19] C. Jie, RSSI-based indoor mobile localization in wireless sensor network, Int. J.
Digit. Content Technol. Appl. 5 (7) (2011) 408–416.
[20] Z. Xiong, F. Sottile, M.A. Spirito, R. Garello, Hybrid indoor positioning
approaches based on WSN and RFID, in: 2011 4th IFIP International
Conference on New Technologies, Mobility and Security (NTMS), IEEE, 2011,
pp. 1–5.
[21] G. Mao, Localization Algorithms and Strategies for Wireless Sensor Networks:
Monitoring and Surveillance Techniques for Target Tracking, IGI Global, 2009.
[22] P. Manual, XBee/xBee-PRO RF modules.
[23] S.S. Saab, Z.S. Nakad, A standalone RFID indoor positioning system using
passive tags, IEEE Trans. Ind. Electron. 58 (5) (2011) 1961–1970.
[24] J. Han, Q. Song, Y. He, Adaptive unscented Kalman filter and its applications in
nonlinear control, in: Kalman Filter Recent Advances and Applications,
InTech, 2009.
289
[25] G. Welch, B.D. Allen, A. Ilie, G. Bishop, Measurement sample time
optimization for human motion tracking/capture systems, Proceedings of
Trends and Issues in Tracking for Virtual Environments, Workshop at the IEEE
Virtual Reality 2007 Conference (2007).
Biographies
Amir Karimi Alavijeh received the M.Sc. degree in Control
Electrical engineering from the University of S. Beheshti,
Tehran, Iran, in 2014. He was a design engineer in sev-
eral companies where he designed power stations, and
worked on research based projects related to image pro-
cessing, energy verifying, power protection, and so on.
He is currently working on developing a wireless sensor
network and researching into localization using wireless
sensor networks and OCR employing neural networks. He
is now a university lecturer at IAU and PNU. His primary
research interests include control engineering, estimation
techniques, robotics, and wireless communication sys-
tems.
Mohammad Hossein Ramezani received the M.S. and
Ph.D. degrees both in Electrical engineering from Sharif
University of Technology, Iran, in 2002 and 2008, respec-
tively. In 2009, he joined the Faculty of Electrical
Engineering, University of S. Beheshti, Iran as assistant
professor. In the period of May 2014–April 2017 he has
been post doctorate fellow at the Faculty of Engineering,
University of Southern Denmark. Since March 2017, he
has been assistant professor of Control Engineering at the
Mads Clausen Institute, Division of Mechatronics, Univer-
sity of Southern Denmark. His primary research interests
are control engineering, mechatronics and signal process-
ing.
Ali Karimi Alavijeh received the B.Sc. degree in Power
Electrical Engineering from Azad University of Najafabad,
Iran, in 2011. He is currently a DSP-based instrument
designer. His research interests include aviation design
and sensor network enhancement (software and hard-
ware).