Investigations on real time RSSI based outdoor target tracking using kalman filter in wireless sensor networks

International Journal of Electrical and Computer Engineering (IJECE)
Vol. 10, No. 2, April 2020, pp. 1943~1951
ISSN: 2088-8708, DOI: 10.11591/ijece.v10i2.pp1043-1951  1943
Journal homepage: http://ijece.iaescore.com/index.php/IJECE
Investigations on real time RSSI based outdoor target tracking
using kalman filter in wireless sensor networks
K. Vadivukkarasi, R. Kumar
Department of Electronics and Communication Engineering, SRM Institute of Science and Technology, India
Article Info ABSTRACT
Article history:
Received Mar 14, 2019
Revised Oct 21, 2019
Accepted Oct 30, 2019
Target tracking is essential for localization and many other applications in
Wireless Sensor Networks (WSNs). Kalman filter is used to reduce
measurement noise in target tracking. In this research TelosB motes are used
to measure Received Signal Strength Indication (RSSI). RSSI measurement
doesn‟t require any external hardware compare to other distance estimation
methods such as Time of Arrival (TOA), Time Difference of Arrival (TDoA)
and Angle of Arrival (AoA). Distances between beacon and non-anchor
nodes are estimated using the measured RSSI values. Position of the non-
anchor node is estimated after finding the distance between beacon and non-
anchor nodes. A new algorithm is proposed with Kalman filter for location
estimation and target tracking in order to improve localization accuracy
called as MoteTrack InOut system. This system is implemented in real time
for indoor and outdoor tracking. Localization error reduction obtained in an
outdoor environment is 75%.
Keywords:
Kalman filter
Localization
Mote track inout system
position estimation
RSSI
Target tracking
Copyright © 2020 Institute of Advanced Engineering and Science.
All rights reserved.
Corresponding Author:
K. Vadivukkarasi,
Department of Electronics and Communication Engineering,
SRM Institute of Science and Technology,
SRM Nagar, Kattankulathur, Kancheepuram, Tamil Nadu 603 203, India.
Email: vadivukk@srmist.edu.in
1. INTRODUCTION
A wireless sensor network is composed of a large number of inexpensive, low-power and tiny
sensors, which are randomly and densely deployed in the surveillance area [1]. These tiny sensors, which
consist of sensing, data processing, and communicating components, collaboratively monitor the event of
interest, process the sensed data, and provide resultant information about the monitored events for a number
of potential applications, ranging from battlefield monitoring and environmental surveillance to health
care [2-5]. Object tracking and localization are the important problems of WSNs [6, 7]. Energy
consumption and accuracy enhancement are the recent concerns in WSN. Localization algorithms are
classified as range based and range free localization. Range free localization does not require distance
estimation. There are four basic range based techniques in WSN, namely AOA, TOA, TDOA and RSSI.
RSSI based localization is the cost-effective implementation because no extra hardware required for
distance measurement compare to other ranging techniques. But the drawback of RSSI based
localization is the relatively lower accuracy. Many localization algorithms have been introduced to improve
the localization accuracy [8-10].
The rest of the paper is organized as follows: Section 2 gives the Literature review. Section 3
presents overview of MoteTrack InOut system and it explains the experimental setup used in this work.
In Section 4 describes about distance and position estimation methods respectively. Section 5 introduces
Kalman filter and section 6 focuses on MoteTrack InOut system algorithm. Section 7 the results from
outdoor environment is provided. Conclusion is provided at the end of paper.

 ISSN: 2088-8708
Int J Elec & Comp Eng, Vol. 10, No. 2, April 2020 : 1943 - 1951
1944
2. RELATED WORK
In [11], an enhanced version of Disk Overlapping localization scheme is proposed by attaching mass
coefficients to each point of the possible locations in the region. The position estimation is derived in [12],
as weighted least square problem and the location estimates are computed in a decentralized manner.
Location estimation may also be defined in a dynamic scenario with the help of location tracking of a single
mobile node (MN). With the information provided by previous locations and the mobility of nodes enable us
to gain better estimation accuracy and a smoother trajectory in this approach of location tracking. Kalman
Filter (KF) has been widely utilized for modeling the tracking of mobile nodes in the literature.
A Unified Kalman Tracking (UKT) algorithm is introduced in [13]. As this algorithm combines
the previously proposed two-step least square method into KF state formulation and provides a simultaneous
solution to the location estimation and tracking problems. Considering RSS variations, [14] has the SNRs at
sensor nodes by a 2nd order polynomial function for their specific experimental setup. Later, KF applied to
the predicted SNRs to track position changes. In [15] the tracking task of a mobile node is also modeled by
KF. Lateralization with nearest three estimated nodes gives advantage to estimate the coordinates of
the object. This has been experimented for straight or circular movements of the mobile nodes. Another
proposed method in [16] has improved the performance of KF by renovating the estimated velocity in state
equations through a direction finding process. In this paper, an adaptive Mote Track InOut tracking technique
is developed to track the variations of the location estimations obtained. The initially modeled KF has been
modified to account for the convergence requirement, uncertainties about system dynamic model and noise
description. The proposed adaptive Kalman tracking scheme can well adapt to the rapid changes of
movement and environment [17-19].
3. EXPERIMENTAL SET UP OF MOTE TRACK INOUT SYSTEM
In an indoor environment, for a moving target the localization error using Log Normal Shadowing
Model (LNSM), Kalman filter and proposed method is 2.231m, 1.007m and 0.818m respectively.
The localization error is reduced using proposed method but over lapping occurred in the tracking path.
MoteTrack InOut is used and it is found that localization error is 0.570m and there is no overlapping in
tracking path. Localization error reduction obtained using the proposed method along with the Kalman filter
for a moving target is 1.66m when compared to LNSM [20]. The outdoor experiments are conducted in an
open play ground. Ground is surrounded by number of trees and plants. Figure 1 show the experimental test
bed set up of MoteTrack InOut system. It consists of four beacon nodes and one mobile node.
Figure 1. The experimental test bed
As a node TelosB mote is employed. It is a low power embedded device with chipcon
CC2420 [21, 22]. It provides a digital received signal strength indicator (RSSI) that may be read any time.
TelosB mote has a USB port, applying that it is connected to PC or laptop. Operating system for this mote is
Tiny OS. We can develop a program in PC or laptop applying NesC language and upload a program into
the TelosB mote as well as erase a program and load a new program. Beacon node will send a empty packet
periodically to the mobile node. This periodic message consists of two information‟s such as source ID and

Int J Elec & Comp Eng ISSN: 2088-8708 
Investigations on real time RSSI based outdoor target tracking using kalman filter… (K. Vadivukkarasi)
1945
power level. There are eight power levels used in TelosB mote. The highest and default power level is 31 and
its equivalent value in dBm (decibel-mill watt) and mill watts is 0 and 1 respectively. The beacon nodes are
fixed at the four places of the ground with coordinates (0, 1), (0, 2.5), (1, 7.1), (2.5, 7.1). The moving target
node is carried by a person. The walking target traverses the corridor at a varying velocity. The RSSI
measurements from the anchor nodes are collected and forwarded to the sink node. The sink node is
connected to PC where the localization algorithm is implemented.
4. DISTANCE AND POSITION ESTIMATION
Log normal shadowing model (LNSM) is used to estimate distance between beacon node and
unknown node. The measured RSSI values are used in LNSM to find distance [23, 24].
      L t rP d P dBm P d dBm  (1)
PL - Total path loss in dB; Pt - Transmitted power in dBm; Pr – Received power in dBm. Propagation model
used in wireless sensor network is given by
     0 010 /L LP d P d n log d d X   (2)
Where „PL(d0)‟ is the path loss at the reference distance d0 in dB, For this experiment „d0‟is taken as 1m , „d‟
is the distance from sender, „n‟ is the path loss exponent, „Xσ‟ is the zero-mean Gaussian random variable.
Path loss exponent measures the rate at which the RSS decreases with distance, and its value depends on
the specific propagation environment
   10rP d A n log d  (3)
Where A = Pt - PL(d0) is the received signal at 1m distance. Linear regression analysis is used to compute ‘A‟
and σ value based on the measured data. The RSSI values are collected at the target node which is at known
distances away from the Beacon nodes. Crammer‟s rule can be used to find the values of A and n.
T=(MT
M)-1
MT
R (4)
[ ], [ ], [ ]
Where d1, d2, d3, d4 represent the distances between the target and the respective beacon nodes and RSSI1,
RSSI2, RSSI3, RSSI4 are the RSSI mean values collected from the beacon nodes.The below equation is used
to calculate distance from RSSI measurements.
(5)
The above equation is used to calculate distance from RSSI measurements. Where „n‟-Path loss
exponent; Pr – Received power in dBm; Multilateration is implemented to find the position of a unknown
node. Four beacon nodes positions are kept at reference to find the unknown node‟s position.
The multilateration method has been selected due to its good computational cost-accuracy trade-off [25].
If the beacon nodes are located at (x1, y1), (x2, y2), (x3, y3), (x4, y4) then the position of the target (x, y) can be
obtained using (6) as
2 2 2
( ) ( )1 1 1
2 2 2
( ) ( )2 2 2
2 2 2
( ) ( )
d x x y y
d x x y y
d x x y yn n n
   
   
   
(6)

 ISSN: 2088-8708
1946
5. KALMAN FILTER
The objective of the Kalman filter is to minimize the mean square error between the actual and
estimated data. It estimates the states of a process in two steps, predictor state and corrector state [26, 27].
Predictor state: This state calculates an estimate of the future state of the system from the previous state
estimate.
Update expected value of X Xk=AXk-1+BU (7)
Update error covariance matrix
Pk=APk-1AT
+Q (8)
Where, Xk is state estimate at step k [ ]
A (n×n) matrix that relates the step of k-1 to current state k
A is taken as [ ], represents the sampling period, B(n×1) matrix that relates U to the state
X , B is taken as [ ], Where „U‟ is the optional control input; „Pk‟ is the error covariance; „Q‟ is
the process noise covariance.
Corrector (Measurement update) state: First step loin this state is calculation of Kalman gain. The second
step is to update expected value and the final step is updation of error covariance matrix.
Compute Kalman gain Kk = Pk-1HT
(HPk-1HT
+R)-1
(9)
Update expected value Xk = Xk-1+Kk(Zk-HXk-1) (10)
Update error covariance Pk = (1-KkH)Pk-1 (11)
H(m×n) matrix that relates the state Xk to the measurement Zk by Zk = HXk + G (12)
Measurement noise covariance is represented by random variable „G‟. Q and G are assumed to be
independent (of each other), and with normal probability distributions. In practice, the process noise
covariance and measurement noise covariance matrices might change with each time step or measurement,
however here we assume they are constant. The time update projects the current state estimate ahead in time.
The measurement update adjusts the projected estimate by an actual measurement at that time. The filter
works in a recursive manner. After each time and measurement update, the steps are repeated with
the previous posterior estimates used to project or predict the new a priori estimates. The Kalman filter
recursively generate a current estimate based on all of the past measurements.
6. MOTE TRACK IN-OUT SYSTEM
Flow diagram of target tracking using MoteTrack InOut system is shown in Figure 2. TelosB motes
are used to measure real time RSSI values in an outdoor environment. After distance and position estimation
using these measurements, position error is calculated. In order to reduce position error a new algorithm is
developed. Kalman filter is applied after this algorithm to track the target and to reduce the measurement
error. The MoteTrack InOut system works upon the following algorithm.
1) Collect RSSI values from beacon nodes and calculate the average of RSSI measurements of each
beacon node. Find path-loss exponent „n‟and „A‟.
2) Estimate distance from log normal shadowing model
3) Find distance error (DE).
DE =Actual distance ( – Calculated distance
4) Find average distance error (ADE).

1947
5) Average distance error ADE is negative value,
a. If the distance error (DE) is positive, ADE will be added to DE. Then the Resultant error
reduction(RER) is
RER = DE + ( - ADE )
b. If DE is negative, ADE will be subtracted from DE. Then the RER is
RER = - DE – (-ADE).
6) ADE is positive value,
a. If DE is negative, ADE will be added to DE. Then the RER is
RER = -DE + ADE
b. If DE is positive, ADE will be subtracted from DE.
RER = DE - ADE.
7) If RER is greater than the DE value, then DE is considered as RER. Calculate average distance error
reduction (AER).
8) Calculate the percentage of error reduction from AER and ADE.
9) Find node‟s position and position error
∑ √( ) ( )
is the calculated position and (xk, yk) is the actual position of the target at kth
position.
When position error ( ) is minimum better performance of the localization system is achieved.
10) Repeat steps from 1 to 9 for various beacon positions.
11) Implement Kalman Filter
a. Initialize the matrices A, B, C of the state equation
b. Set the initial velocity to zero and set the initial position
c. for k = 1 to 25 do
i. Xk=AXk-1+BU
ii. Pk=APk-1AT
+Q
iii. Kk=Pk-1HT
(HPk-1HT
+G)-1
iv. Xk=Xk-1+Kk(Zk-HXk)
v. Pk=(1-KkH)Pk-1
d. end for
12) Plot the localization graph.
This algorithm is used to reduce the position error in indoor and outdoor environments.
Figure 2. Flow diagram of motetrack inout system
7. TARGET TRACKING COMPARED WITH OTHER METHODS
Kalman filter is widely used in control systems to estimate the state of a process in presence of noisy
measurements. It estimates the states of a process by minimizing mean square error between the ideal and
real system states.

 ISSN: 2088-8708
1948
7.1. Outdoor Environment System Model
Table 1 shows the parameters involved for the RSSI measurements in an outdoor environment.
Four anchor nodes are used for this experiment and it is named as N1, N2, N3 and N4. Readings are taken in
128 different positions using non-anchor node. From Figure 3 to Figure 6 show the final simulation results of
the localization system. The result compares the actual path with the estimated path. Right subplot of
Figure 3 represents the actual path followed by the moving target. Every measurement was carried out on
the marked points (blue colour square boxes). Anchor node positions are shown in red colour small circle.
Left subplot of Figure 3 shows the tracking without applying the filter. Here the tracking is done only using
log-normal shadowing model (LNSM). The average position error per position is 3.425m. As shown in
Figure 3, the obtained path doesn‟t draw the actual path. Some of the results go beyond the boundaries.
Figure 4 compares the actual path with the one obtained after Kalman filter is implemented.
Left subplot of Figure 4 shows the path obtained after filtering the positions. The Kalman Filter smoothen
the path as well as reduces the error. The error in the estimation of the target is reduced to 2.112m.n Figure 5
compares the actual path with the one obtained after proposed method before applying Kalman filter is
implemented. Left subplot of Figure 5 shows the path obtained after filtering the positions. The proposed
method reduces the error as well as smoothen the path but in some places overlapping occurred. The error in
the estimation of the target is reduced to 1.251m.
Table 1. Real time outdoor parameters
Number of Anchor
nodes used
Anchor node
positions(N1,N2,N3,N4)
Number of mobile
nodes used
Mote Number of
positions used
Operating
System
Simulation
4 (0,1),(0,2.5),(1,7.1),(2.5,7.1) 6 TelosB 128 TinyOS MATLAB
Figure 3. Plot comparing the actual path with estimated path obtained using LNSM
Figure 4. Plot comparing the actual path with estimated path using kalman filter
-5 0 5 10
-1
0
1
2
3
4
5
6
7
8
x position in m
ypositioninm
tracking with LNSM
0 1 2 3 4
0
1
2
3
4
5
6
7
8
x position in m
ypositioninm
actual path
0 2 4 6
2
3
4
5
6
7
8
9
10
x position in m
ypositioninm
tracking with kalman filter
0 1 2 3 4
0
1
2
3
4
5
6
7
8
x position in m
ypositioninm
actual path

1949
Figure 5. Plot comparing the actual path with estimated path obtained using proposed method before
applying kalman filter
Figure 6 compares the actual path with the one obtained after MoteTrack InOut is implemented.
It is found that localization error is 0.834m and there is no overlapping in the tracking path. The Localization
errors obtained using the proposed method and various other methods are shown in the Table 2.
Figure 6. Plot comparing the actual path with estimated path obtained using motetrack inout system
Table 2. Localization error calculated for the real time RSSI values under different methods
Method Localization Error((m)
LNSM 3.425
Kalman filter 2.112
Proposed method without Kalman filter
MoteTrack InOut System
1.251
0.834
8. CONCLUSION
In an outdoor environment, for a moving target the localization error using LNSM, Kalman filter
and proposed method before implementing Kalman filter is 3.425m, 2.112m and 1.251m respectively.
The localization error is reduced using proposed method but over lapping occurred in the tracking path.
Proposed algorithm with the Kalman Filter is used and it is found that localization error is 0.834m and there
is no overlapping in tracking path. Localization error reduction obtained using the MoteTrack InOut system
for a moving target is 2.51m when compared to LNSM. Localization error reduction of 74.5% and 75.6% are
obtained for indoor and outdoor respectively using Motetrack InOut system for a moving target compared to
LNSM.
0 1 2 3 4
0
1
2
3
4
5
6
7
x position in m
ypositioninm
tracking with proposed method
0 1 2 3 4
0
1
2
3
4
5
6
7
8
x position in m
ypositioninm
actual path
0 1 2 3 4
0
1
2
3
4
5
6
7
8
x position in m
ypositioninm
tracking with kalman filter & proposed method
0 1 2 3 4
0
1
2
3
4
5
6
7
8
x position in m
ypositioninm
actual path

 ISSN: 2088-8708
1950
REFERENCES
[1] Akyildiz, W. Su, Y. Sankarasubramaniam, and E. Cayirci, "Wireless sensor networks: A survey," Computer
Networks Journal, vol. 38(4), pp. 393-422, 2002.
[2] D. Estrin, L. Girod, G. Pottie and M. Srivastava, "Instrumenting the world with wireless sensor networks,"
International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2001), pp. 2685-2678, 2001.
[3] C. Otto, A. Milenkovic, C. Sanders and E. Jovanov, "System architecture of a wireless body area sensor network
for ubiquitous health monitoring," Journal of Mobile Multimedia, vol. 1(4), pp. 307-326, 2006.
[4] Alaa jabber, Robert Bicker, “Design of a Wireless Sensor Node for Vibration Monitoring of Industrial Machinery”,
International Journal of Electrical and Computer Engineering (IJECE), Vol. 6((5), pp. 639-653, April 2016
[5] Fatma Karray, Mariem Triki, Mohamed Wassim Jmal , Mohamed Abid, and Abdulfattah M. “WiRoTip: an IoT-
based Wireless Sensor Network for Water Pipeline Monitoring,” International Journal of Electrical and Computer
Engineering (IJECE), Vol. 8, No. 5, pp. 3250 – 3258, October 2018.
[6] Huang, X. “Target Localization Based on Improved DV-Hop Algorithm in Wireless Sensor Networks,” Journal of
Networks, vol 38(4), pp.393-422, 2014.
[7] Nguyen, T. L. N., Shin, Y. “Multiple Target Larget Localization in WSNs based on Compressive Sensing Using
Deterministic Sensing Matrices,” International Journal of Distributed Sensor Networks, 16, pp.1-8, 2015.
[8] Pal, A, “Localization Algorithms in Wireless Sensor Networks:Current Approaches and Future Challenges.
Network Protocols and Algorithms,” Network Protocols and Algorithms, 2(1), pp. 45-73, 2010.
[9] Yao, Y. “A RSSI-Based Distributed Weighted Search Localization Algorithm for WSNs,” International Journal of
Distributed Sensor Networks, 29(3), pp.1-11, 2015.
[10] Cui, F. “Weighted Centroid Localization Algorithm based on Multilateral Localization Error of Received Signal
Strength Indicator,” Infrared and Laser Engineering, 44(7), pp. 2162-2168, 2015.
[11] Mahdy Nabaee, Ali Olfat, "Enhanced RSS based location estimation with disk overlapping in wireless sensor
networks," IEEE International Conference on Telecommunications, 2008.
[12] Chin-Liang Wang, Yih-Shih Chiou, Yu-Sheng Dai, "An adaptive location estimator based on kalman filtering for
wireless sensor networks," IEEE 65th International Conference on Vehicular Technology, 2007.
[13] Chin-Liang Wang, Yih-Shih Chiou, "An adaptive positioning scheme based on radio propagation modeling for
indoor WLANs," IEEE 63rd International Conference on Vehicular Technology, 2006.
[14] Jamin Lee, Hyungjoon Song, Sungmok Oh, Daesik Hong, "An improved location tracking algorithm
with velocity estimation in cellular radio networks," IEEE 63rd International Conference on
Vehicular Technology, 2006.
[15] Po-Hsuan Tseng, Chao-Lin Chen, Kai-Ten Feng, "A unified kalman tracking technique for wireless location
systems," 2nd International Symposium on Wireless Pervasive Computing (ISWPC '07), 2007.
[16] S. Umesh Babu, C. S. Kumar, R. V. Raja kumar, "Sensor networks for tracking a moving object using kalman
filtering," IEEE International Conference on Industrial Technology (ICIT), 2006.
[17] Giovanni Zanca, Francesco Zorzi, Andrea Zanella and Michele Zorzi, "Experimental comparison of RSSI-based
localization algorithms for indoor wireless sensor networks," Proceedings of the workshop on Real-world wireless
sensor networks, pp. 1-5, 2018.
[18] José Ramiro Martínez-de Dios, Anibal Ollero, Francisco José Fernández and Carolina Regoli, "On-line RSSI-range
model learning for target localization and tracking," Journal of Sensor and Actuator Networks, vol. 6(15),
pp. 1-19, 2017.
[19] Sadik Kamel Gharghan, Rosdiadee Nordin, Aqeel Mahmood Jawad, Haider Mahmood Jawad, and Mahamod
Ismail, "Adaptive neural fuzzy inference system for accurate localization of wireless sensor network in outdoor and
indoor cycling applications," IEEE Access, vol. 6, pp. 38474-38489, 2018.
[20] K. Vadivukkarasi, R. Kumar, Mary Joe, "Real time RSSI based novel algorithm to improve indoor localization
accuracy for target tracking in wireless sensor networks," ARPN journal of engineering and applied sciences,
vol. 10(16), pp. 7015-7023, 2015.
[21] TelosB mote platform datasheet, http://www.willow.co.uk/TelosB_Datasheet.pdf.
[22] "CC2420 Radio Data Sheet Revision 1.3." http://www.chipcon.com/files/CC2420 Data Sheet 13.pdf, October 2005.
[23] T. S. Rappaport, "Wireless communications: principles and practice", 2nd edition, Prentice Hall, 2002.
[24] Chuan-Chin Pu, and Wan-Young Chung, "Mitigation of multipath fading effects to improve indoor RSSI
performance," IEEE Sensors Journal, vol. 8(11), pp 1884-1886, 2008.
[25] Jokhio S. H., Jokhio I. A., Kemp A. H., "Light-weight framework for security-sensitive wireless sensor networks
applications," Wireless Sensor Systems, IET, vol. 3(4), pp. 298-306, 2013.
[26] Abdalkarim A wad, Thorsten Frunzkeand Falko Dresslerr, "Adaptive distance estimation and localization in SN
using RSSI measures," IEEE 10th Euromicro Conference on Digital System Design Architectures, Methods and
Tools, pp. 471-478, 2007.
[27] Pashazadeh S, Sharifi M., "A geometric modelling approach to determining the best sensing coverage for
3-dimensional acoustic target tracking in wireless sensor networks," Sensors journal, vol. 9(9), pp. 6764-6794,
2009.

1951
BIOGRAPHIES OF AUTHORS
K. Vadivukkarasi received her PhD degree from SRM University, India (2016), Master of
Engineering degree in Embedded System Technology from Anna University, Tamilnadu, India in
2007 and Bachelor of Engineering degree in Electronics and Communication Engineering from
Madras University, Tamilnadu, India, in 2001. She is working an assistant professor in
the Department of Electronics and Communication Engineering, SRM University, India. Her main
research interests are in wireless sensor networks, Embedded system design, Embedded
networking and Internet of Things.
R. Kumar received his PhD degree from SRM University, India (2009), Master of science from
BITS Pilani (1993) and bachelor‟s degree in Electronics and Communication Engineering from
Bharathidasan University, 1989. He is working as a professor in the Department of Electronics and
Communication Engineering, SRM University, Chennai, India. His areas of interest include spread
spectrum techniques, wireless communication, wireless sensor networks MIMO–OFDM systems
and under water fiber optic communication. He has supervised a number of undergraduate,
postgraduate as well as PhD students in the area of wireless communication. His research work led
him to publish more than 100 papers in journals and in conferences.

Investigations on real time RSSI based outdoor target tracking using kalman filter in wireless sensor networks

More Related Content

What's hot (16)

Similar to Investigations on real time RSSI based outdoor target tracking using kalman filter in wireless sensor networks (20)

More from IJECEIAES (20)

Recently uploaded (20)

Investigations on real time RSSI based outdoor target tracking using kalman filter in wireless sensor networks