Content uploaded by Phaklen Ehkan
Author content
All content in this area was uploaded by Phaklen Ehkan on Sep 06, 2016
Content may be subject to copyright.
Hardware Comparison Capturing Received Signal
Strength Indication (RSSI) for Wireless Sensors
Network (WSN)
M.I. Jais1, P. Ehkan6, R.B. Ahmad7, I. Ismail8
1,4,5,6Embedded, Network and Advance Computing (ENAC)
School of Computer and Communication Engineering
Universiti Malaysia Perlis
Pauh Putra, Perlis, Malaysia
ilmanjais86@gmail.com1, phaklen@unimap.edu.my,6,
badli@unimap.edu.my7, iszaidyismail.gmail.com8
Abstract— Wireless sensor network (WSN) is commonly used for
localization analysis. Through sniffing receive signal strength
indication (RSSI) in WSN system, localization of higher RSSI is
connected automatically. This paper proposes the Raspberry-Pi
(RasPi) based sensor node to sniffing RSSI and automatically
connect the device to strongest RSSI. The RasPi brings the
advantages of a personal computer (PC) to the domain of sensor
network which makes it as the perfect platform for interfacing with a
wide variety of external peripherals. Comparative analysis of its key
elements and performances with some of current existing wireless
sensor nodes has shown that despite the few disadvantages, the RasPi
remains an inexpensive single board computer (SBC) with its very
successfull use in sensor network domain and diverse range of
research applications.
Keywords— Wireless sensor network (WSN), Raspberry-Pi (RasPi),
Received Signal Strength Indication (RSSI)
I. INTRODUCTION
A WSN is a network of devices that have sensing,
actuation, processing, and wireless communication capability.
Wireless sensor networks have several potential applications,
especially in the area of transportation operations[1]–[4].
Sensor scans are used to detect physical phenomena, such as
motion and acceleration. The collected and processed sensor
data can be transmitted through wireless links. Position
monitoring or also known as location estimation is the process
of obtaining location of a sensor node. The effective use of the
data from the sensors, position of the sensors must be known.
Finding the location of the nodes manually is costly and
T.Sabapathy2, M. Jusoh3, H.A. Rahim4,
2,3Radio Engineering Research Group (RERG)
School of Computer and Communication Engineering
Universiti Malaysia Perlis
Pauh Putra, Perlis, Malaysia
thenna84@gmail.com2, muzammil@unimap.edu.my3,
haslizarahim@unimap.edu.my4
sometimes it is not feasible. Therefore, nodes should estimate
their own locations through several methods. WSN is
composed of spatially distributed nodes equipped with sensing
devices to monitor and to measure characteristics of the
physical environment at different locations. WSNs are
designed and deployed for different purposes by various
organizations. WSN based monitoring applications range from
simple data gathering, to complex Internet-based information
systems. In other words, the observations obtained from
sensor networks may be helpful in many software applications
like environmental, industrial and meteorological monitoring,
building and home automation, medicine, urban sensor
networks, intelligent transportation, security, military defense,
etc [1]. Sensor nodes are the small, low power single board
computers with a radio for wireless communication. Number
and types of sensors depend on the applications. Sensor nodes
collect and transfer data using four stages: collecting the data,
processing the data, packaging the data and communicating
the data [5]. In this work, the main objective is to observe
abilities of SBC especially Ras-Pi to perform as RSSI sensor
based of centre processing unit (CPU) and memory
performance.
The remainder of this paper is organized as follows. In
section II, the overview of overall system architecture.
Followed in section III, the experiment methodology are
described. In section IV, significant result are review and
discussed. Finally, conclusion this paper in section V.
M.F. Malek5
University of Wollongong in Dubai
Faculty of Engineering and Information Sciences
Knowledge Village Dubai
United Arab Emirates
mohamedfareqmalek@uowdubai.ac.ae5
2015 IEEE Student Conference on Research and Development (SCOReD)
278
II. SYSTEM ARCHITECTURE
A unique design and implementation of a secure sensor
node has been carried out based on three major components: a
single board computer, wi-fi adapter module and various
routers as an access point (AP). In this section, overview of
the hardware architecture and software involves in this
proposed experimental.
A. Raspberry-Pi
RasPi is education oriented pocket size single board
computer (SBC) with small size, inexpensive and hackable.
Due to its low price and capability to be as aperfect platform
for interfacing with many devices, theRasPi has been adopted
as sensor node to process the RSSI information.
Fig. 1. Raspberry. (a) Model Pi2. (b) Model B+
The RasPi board consists of a processor and graphics chip,
program memory (RAM) and various interfaces and
connectors for external devices . All RasPi models have the
similar CPU named BCM2835 which is low cost, low power
consumption and it is powerful [5]–[7]. Figure 1 shows the
two types of RasPi model. Both of this model is used in this
work to implement the sensor node system. In similar way
like a standard PC, RasPi with a keyboard for command entry,
a display unit and a power supply. Mini-SD Flash memory
card is configured in such a way to represent a hard drive to
RasPi processor. The unit is powered via the micro USB
connector. Internet connectivity is set through either via an
Ethernet/LAN cable or via a USB dongle (WiFi connectivity)
[8], [9]. Like any other computer, the RasPi also uses an
operating system and the “stock” OS is a flavor of Linux
called Raspbian. Linux, as a free and open source program, is
a great match for RasPi. Since it is free, it keeps the price of
the platform low.On the other hand, it makes the RasPi more
hack-able. There are also a few non-Linux OS options
available [8], [9]. The additional hardware and software
requirements can be achieved by already existing hardware
modules and open source software. One of the great things
about the RasPi is that it has a wide range of usage.
Numerous method of RasPi usage with sensors, displays
and motors is given in [10]. In the workshop [11], a RasPi is
used to read sensors (inputs), store their values in a database
for historical trending and turn relays (outputs) on and off
when a sensor value goes outside of a certain range. The
specification of RasPi is shown in Table I.
TABLE I. Comparison Experimental Platform
B. Communication Devices
A key evaluation metric for any WSN is its
communication rate, power consumption, and range. In order
to create an interconnected network on real scenarios, various
nodes be placed far apart and randomly.
Fig. 2. TP-Link 54 Mbps Wireless Adapter
The Ethernet port is the RasPi main gateway for
communication with other devices and the Internet. However,
wireless adapter TP-Link in Figure 2 is deployed with RasPi-
for this experiment. Wireless adapter attached with RasPi
have 54 Mbps. Meanwhile, for another competitor platforms
deploy, build in wireless adapter as shown in Table I. All
WiFi adapters will connect with multiple access point.
Name Processor Memory
Operating
System
Wi-Fi
Module
Dell
Inspiron
Intel(R)
Core(TM) i5-
4210
(2.4GHz)
4GB
Ubuntu
14.04 LTS
64 Bits
Intel (R)
Dual
Band
Wireless-
AC 3160
Lenovo
Intel ATOM
Processor
N270 (1.6
GHz)
1GB
DDR2-667
SDRAM
SO-DIMM
(PC2-5300)
Ubuntu
12.04 LTS
32-Bits
Bluetooth
with
Enhanced
Data Rate
(BDC-
2.1)
Raspi-B+
ARMv6
Single-Core
(700MHz)
512 MB
SDRAM @
400 MHz
Raspbian
TP-Link
TL-
WN722N
Raspi- 2
ARMv7
Quad-Core
(900MHz)
1 GB
SDRAM @
400 MHz
Raspbian
TP-Link
TL-
WN722N
(b)
(a)
2015 IEEE Student Conference on Research and Development (SCOReD)
279
C. Software Structure
The Python version 2.7 is adopted to program the sensor
node. Python is one of common use programming high-level
language. Python is used to to program the RSSI detection
and selection based on th highest RSSI. A detailed
information of the programming is presented in the next
section.
III. EXPERIMENTAL METHODOLOGY
The flow chart depicted in Figure 3 demonstrates the
operation of the proposed system. Based on this flow chart,
before the capturing starts, the user needs to scan all incoming
RSSI. Each platform absorbs any incoming RSSI from the
various router for (tens) 10 times as demonstrated in Figure 4.
Simultaneously, the RSSI value is sort out according to the
maximum value and after capturing RSSI process end, the
WiFi module is connected to the higher incoming RSSI value.
This process is repeated for all experimental platforms. Each
platform has differences architecture as shown in Table I. The
main reason of this research work is to prove that RasPi can
perform as a sensor node in WSN for localization purpose.
Fig. 3. Overall Flowchart of RSSI capturing system
Fig. 4. Topography of RSSI capturing system.
IV. SYSTEM PERFORMANCE EVOLUTION AND
DISCUSSION
Fig. 5. Screenshot of RSSI value of Ubuntu through Python 2.7
Fig. 6. Screenshot of RSSI value on Raspbian through Python 2.7
2015 IEEE Student Conference on Research and Development (SCOReD)
280
A. CPU result (between Board)
Fig. 7. CPU Performances during Sniffing RSSI Value
CPU performance (%) = 100 % – CPU idle % (1)
CPU utilization analysis is shown in Figure 7. As can be
observed in Figure 7, all types of platforms finished scanning
and capturing RSSI below than 30 seconds. However,
platform Dell and RasPi-2 show a lowers CPU usage
compared than other platform. Meanwhile, RasPi-B+ used the
highest CPU usage because it's internal hardware
specification, formerly lower compared than other platforms.
However, RasPi-B+ still compatible with running a sensor
node to capturing incoming RSSI. The unstable graph
occurred due to the background process which runs
simultaneously.
B. Memory result (between Board)
Fig. 8. Memory Usage during Sniffing RSSI Value
Memory performance (%) = (Memory performance / Total
memory – free space) x 100% (2)
Fig. 8 illustrates the memory used for sensoring RSSI by
every platform. Average memory used by Dell and RasPi-2 is
around 20% - 30%. Therefore, Lenovo netbook and RasPi-B+
used 58% and 68% respectively. Although Lenovo netbook
and RasPi used more than 50% memory caused by other
process, but during capturing RSSI actually it does not
influence percentage of the memory usage. Surprisingly,
RasPi-2 takes less memory usage compared than memory
usage compared than others platform. Hence, it is evident that
RasPi-2 can perform as RSSI capturing and processing
incoming RSSI in WSN system.
V. CONCLUSION
In this work, we have designed a secure sensor node
prototype at a low cost. The RasPi is the best choice because
of its high performance, affordable with the cost, good
memory capacity and being the cheapest single board
computer available in the market as discussed in section II.
Furthermore, the Linux operating system usage provides
additional advantages of using RasPi as a SensorWeb node.
Programming in high-level language, the implemented
solution is quite simple and it is enabled for a large number of
users, as opposed to micro controller programming which
usually depends of development kit. Overall, the
developedplatform successful in capturing and sorting
incoming RSSI in 30 Seconds. The performance evaluated in
terms of CPU and Memory usage shows the credit card size
computer is competent to act as WSN because ease to hack-
able and mobility. However, background process of all
platforms during capturing RSSI, sometimes affect the
performance of platforms, but this problem can be
encountered by terminating unwanted background process.
REFERENCES
[1] V. Vujović and M. Maksimović, “Raspberry Pi as a Wireless Sensor
Node: Performances and Constraints,” Inf. Commun. Technol.
Electron. Microelectron. (MIPRO), 2014 37th Int. Conv., no. May,
pp. 1247–1252, 2014.
[2] S. Banerjee, D. Sethia, T. Mittal, U. Arora, and A. Chauhan,
“Secure sensor node with Raspberry Pi,” IMPACT 2013 - Proc. Int.
Conf. Multimed. Signal Process. Commun. Technol., pp. 26–30,
2013.
[3] H. Zhang, J. Zhang, and H. Wu, “AN ADAPTIVE
LOCALIZATION ALGORITHM BASED,” pp. 2–5, 1857.
[4] P. G. Student, “Wireless Sensor Network Communication Terminal
Based on Embedded Linux and Xbee,” pp. 1468–1473, 2014.
[5] A. Paramanathan, P. Pahlevani, S. Thorsteinsson, M. Hundebøll, D.
E. Lucani, and F. H. P. Fitzek, “Sharing the Pi: Testbed
Description and Performance Evaluation of Network Coding on the
Raspberry Pi,” 2014.
[6] A. M. Damle, “A Low-Cost Embedded Platform for Synchronised
Wide Area Frequency Measurement.”
[7] M. Kochláň, M. Hodoň, L. Čechovič, J. Kapitulík, and M. Jurecka,
“WSN for Traffic Monitoring using Raspberry Pi Board,” vol. 2,
pp. 1023–1026, 2014.
2015 IEEE Student Conference on Research and Development (SCOReD)
281
[8] G. S. Sundaram, B. Patibandala, H. Santhanam, S. Gaddam, V. K.
Alla, G. R. Prakash, S. C. V. Chandracha, S. Boppana, and J. M.
Conrad, “Bluetooth communication using a touchscreen interface
with the Raspberry Pi,” Conf. Proc. - IEEE SOUTHEASTCON,
2013.
[9] E. Hod, A. W. Is, and R. Pi, “WiFiPi-Tracking at Mass Events,”
vol. 00, no. c, 2015.
[10] K. Raguvaran, “Raspberry PI Based Global Industrial Process
Monitoring Through Wireless Communication,” no. February,
2015.
[11] B. Deshmukh, “Raspberry Pi for Automation of Water Treatment,”
pp. 1999–2003, 2014.
2015 IEEE Student Conference on Research and Development (SCOReD)
282
