2014 International Conference on Computing, Networking and Communications, Wireless Ad Hoc and Sensor Networks

Symposium

Wireless Sensor Network-based

Air Quality
Monitoring
System
1
1
2
3
Samer Mansour , Nidal Nasser , Lutful Karim , Asmaa Ali

Electrical & Software Engineering Dept., College of Engineering, Alfaisal University, Saudi Arabia
2
School of ICT, Seneca College of Applied Arts and Technology, Canada
3
School of Engineering, University of Guelph, Canada
1
Email: {smansour, nnasser} @alfaisal.edu, 2 [email protected], [email protected]

Abstract This paper proposes a simple Wireless Sensor

Network (WSN)-based air quality monitoring system (WSNAQMS) for industrial and urban areas.
The proposed
framework comprises a set of gas sensors (ozone, CO, and NO2)
that are deployed on stacks and infrastructure of a Zigbee WSN
and a central server to support both short-term real-time
incident management and a long-term strategic planning. This
architecture would use open-hardware open-software gas
sensing capable motes [6] made by Libelium. These motes use
the ZigBee communication protocol and provide a real-time low
cost monitoring system through the use of low cost, low data
rate, and low power wireless communication technology. The
proposed monitoring system can be transferred to or shared by
other applications. We also introduce a simple but efficient
clustering protocol dubbed hereafter Clustering Protocol for
Air Sensor network (CPAS) for the proposed WSN-AQMS
framework. CPAS proves to be efficient in terms of network
energy consumption, network lifetime, and the rate at which
data is communicated.
Keywords Wireless Sensor Network; Air Quaity
Monitoring Systems (AQMS); Gas Sensing Capable Motes;
LEACH; NO2; CO; Ozone.

I.

INTRODUCTION

Air quality monitoring and control has gained much attention

lately because of the increased environmental awareness and
the effect of air quality on many aspects of life. Besides the
detrimental effects of toxic emissions on the environment and
health, work productivity and energy efficiency are affected
by air quality. Many researches have shown that, in a work
place, the rise of CO2 levels leads to a rise in the quantity of
volatile organic compounds (VOCs), odors, and microorganisms in the air, leading to a decrease in the ability of
individuals to concentrate. Moreover, some studies have
shown that CO2-based air controls can result in up to 50%
energy savings (CO2-based ventilation control can typically
reduce HVAC cost in most buildings by 5% to 20%).
Recently, Wireless Sensor Networks (WSNs) have achieved a
great potential for a widespread applicability in the fields of
monitoring, surveillance, data collection, and medical
telemetry [4]. This potential can be attributed to their
attractive characteristics: (i) WSNs can perform selfconfiguration and reconfiguration in the case of any changes
(for example a network topology change) (ii) WSNs can be
monitored remotely (iii) WSNs can be expanded and redeployed easily (iv) WSNs adapt well to mobility. Despite all
these qualities, the usage of WSNs in air quality monitoring
industries has neither been widely adopted nor efficiently
implemented. Potentials of WSNs in air quality monitoring
have not been exploited to their fullest. In fact, industries are
still using mainly traditional systems when it comes to air
quality and pollution monitoring. These traditional systems,
even though they use sensors, do not enjoy the same levels of
flexibility and capabilities as WSNs. Their fixed topologies
and their inability to self-reconfigure like WSNs, does not

978-1-4799-2358-8/14/$31.00 2014 IEEE

allow the same capabilities in online real time monitoring and

alarm notifications. Hence, timely precautions cannot always
be taken for example as in the case of any incident that forces
some changes in the network topology.
Some WSN-based air quality monitoring systems have been
introduced recently but they are not appealing enough to
industry. Most of these are too complicated to implement,
require specific equipment that is not open-hardware or opensoftware, and are application and location dependent. They
do not consider quality of service (QoS) metrics of the
networks such as delay, accuracy, reliability (fault tolerance).
Data gathered from these WSNs cannot be reused or shared.
Here we introduce a WSN-based air quality monitoring
system for urban and industrial areas. This monitoring
framework uses open hardware and software based on
Libeliums gas sensing capable motes [11]. A set of gas
sensors measuring ozone, CO, and NO2 concentrations is to
be deployed on stacks and infrastructure. Sensed data will be
sent to a dynamically configurable computing platform that
scales to support both near real-time incident management and
longer term strategic planning decisions. Collected data will
be delivered to a computing platform where it is to be
comprehensively analyzed and used to validate the findings of
the existing predicted models. Thus, the proposed framework
provides timely distributed measurements of different air
quality metrics which would help in evaluating the impact of
industrial emissions.
The following features of the proposed framework
distinguish it from the existing air quality monitoring systems
[2 - 10].
An efficient and simple way of event subscriptions and
notifications.
Though it is designed mainly for the air quality
monitoring, the framework can still be reused for other
monitoring applications.
Real time monitoring and alerting on abnormal events
more accurately.
Location and time independent.
An efficient and simple way of identifying the failure of
sensor nodes and reducing sensors energy
consumptions.
Moreover, we introduce a simple Clustering Protocol of Air
Sensor network (CPAS) that fits well with the WSN-AQMS
and that is energy efficient. Simulation results show that the
CPAS protocol has longer network lifetime and can identify
the failure of sensor nodes.

545

2014 International Conference on Computing, Networking and Communications, Wireless Ad Hoc and Sensor Networks
Symposium

The rest of the paper is organized as follows. Section II

presents the preliminary ideas about air quality monitoring.
Section III presents some existing work on AQMS and routing
protocols. Section IV presents the architecture of the proposed
WSN-based air quality monitoring system and CPAS routing
protocol. In Section V we evaluate the performance of the
proposed architecture and routing protocol. Section VI
concludes the paper with some ideas for future research in this
area.
II.

PRELIMINARIES

The Air Quality Index (AQI) is a common indicator of air

quality. The AQI is calculated based on air pollutants such as
CO, SO2, O3, NO2, and other sulfur compounds that have
adverse effects on the environment and human health. The
AQI is a number that represents the highest concentration of a
particular air pollutant at a specific time. Normally, small
concentrations are measured using ppb (parts per billion),
which represents units of mass of a contaminant per one
billion units of total mass. Parts per million (ppm) is a similar
and commonly used unit to measure concentrations of
pollutants. The air quality is considered very good, good,
moderate, poor and very poor for the AQI values, <16, 16-31,
32-49, 50-99, >99, respectively.
Note that for solid and liquid pollutants carried in the air
(known as particulate matter or PM10), concentration is
measured by g/m3. These particles might be smoke, dirt,
mold, pollen etc. PM10 are the larger coarse particles that are
kicked up into the air and spread around causing severe
health effects.
Having come to existence with the birth of the industrial
revolution, the air pollution problem keeps increasing in
severity with the increasing need for power production, and
the increasing number of vehicles. Only recently, within the
last 25 years, that awareness about the problem was raised
worldwide. All nations are concerned and it is a major
problem in middle-eastern countries. For instance, the
Kingdom of Saudi Arabia (KSA) is the 5th air-polluted
country in the world, where the annual mean PM10 is
143g/m3. Moreover, KSA is the 14th in rank for carbon
emission in the world and is responsible for 1.45% of the
world carbon emission. The urban area of KSA accounts for
54% of the pollution.
AQMSs should gain more and more attention since
distributed and localized monitoring of air pollution is really
a major factor in determining the sources and causes of the
pollution and hence in providing solutions. This is true for all
nations especially the middle-eastern ones where, despite the
severity of the problem there, no major steps have been taken
to resolve it.
III.

RELATED WORK

Wireless sensor networks (WSN) provide an excellent, easyto-deploy, and cheap solution to be used in real time automatic
monitoring and control for air quality purposes. Hence,
several WSN-based air quality monitoring systems have been
proposed. Some included sensor networks mounted on cars.
For instance, Mobile Air Quality Monitoring Network
(MAQUMON) [2] is comprised of sensor nodes mounted on

cars, where each sensor node has a set of gas sensors

measuring ozone, CO, and NO2 concentrations. When the car
is in motion, samples are taken very frequently, whereas it is
taken a few times an hour when the car is parked. The samples
are stored and tagged with a location and time information.
When a car is within the coverage area of an available Wi-Fi
hotspot, all data are transferred to the servers, processed and
published on the Sensor Map portal. Vehicular Wireless
Sensor Network (VSN) [3] is another mobile air quality
monitoring system with sensors mounted on cars. A vehicle is
equipped with a CO2 sensor, a GPS receiver, a GSM module,
and a ZigBee-based intra-vehicle wireless network. These
vehicular sensors roam inside the area of interest and
periodically report their sensed data through GSM short
messages. The reported data is collected by a server, which is
integrated with Google Maps to demonstrate the result. An
integrated mobile environmental sensing system [4] was
developed to support the management of transport and urban
air quality. Sensor nodes are deployed on vehicles and
infrastructure to monitor traffic, weather and pollutant
concentrations at far higher spatial and temporal resolutions
and send data into a dynamically configurable computing
platform that supports both near real-time incident
management and longer term strategic planning decisions.
Geo-Sensor networks-based air quality monitoring systems
uses context model [5] to understand the status of pollution in
different areas and provide alarm and safety guidelines.
However, these frameworks are too complicated to implement
and to be reused in other monitoring applications and also are
not convenient and economical for air quality monitoring.
The work done by Jelicic et al. [7] introduces an Indoor Air
Quality monitoring (IAQ) system using a sensor network that
integrates a power management approach to reduce sensors
energy consumption by using an adaptive duty cycling
mechanism for metal oxide semiconductor (MOX) gas
sensors. The work done by Liu et al. [10] proposes a
Wireless Sensor Network (WSN)-based urban air quality
monitoring system that is connected to a GSM system for
centralized control by a LabVIEW program that stores
sensors data in a database. They implemented the monitoring
systems in the city road of Taipei to monitor the carbon
monoxide (CO) concentration caused by vehicle emissions.
However, the proposed system does not deal with any
network management approaches to achieve energy
efficiency. Wireless Sensor Network Air Pollution
Monitoring System (WAPMS) [9] has been implemented to
monitor air pollution in Mauritius. The WAPMS integrates a
new data aggregation algorithm, Recursive Converging
Quartiles (RCQ) that uses the quartile statistical analysis tool
to reduce duplicate data. However, WAPMS uses hierarchical
network topology with single point of failure.
IV. THE PROPOSED FRAMEWORK
In this section, we present the architecture of the proposed
AQMS along with the Clustering Protocol for Air Sensor
network (CPAS). See Figure 1.
A. Architecture of the Proposed Air Quality Monitoring
System
The different components of the CPAS are presented in what
follows.

546

2014 International Conference on Computing, Networking and Communications, Wireless Ad Hoc and Sensor Networks
Symposium

Wireless Sensor Network (WSN)

The proposed WSN-based AQMS uses the Waspmote made
by Libelium [11], which is an open-hardware open-software
wireless sensor device geared towards developers. A single
Waspmote unit can support at the same time one or more of
the following protocols: Zigbee, Bluetooth, RFID, Wi-Fi,
3G/GPRS. It supports connecting two radios at the same time
using an expansion radio board. Thus, a lot of different
combinations are possible. The Wi-Fi module supports direct
connections with iPhone and Android smart phones and also
with any standard Wi-Fi router. The Waspmote has 2 GB
internal storage and can hold a large amount of sensed data.
Over 60 different sensors are available for Waspmote. It has
built-in temperature and accelerometer sensors. Waspmote
supports a gas sensor board that can integrate many gas
sensors (CO, CO2, H2, NO2, CH4, NH3, O3 ) and can also
measure gases concentration levels in the air.
Libelium provides a mesh gateway router, Meshlium, which
provides a connection to the outside world. Meshlium reads
the sensor frames coming from the WSN nodes and can store
them in its internal data base and in external cloud systems
located on the Internet. It supports 802.15.4/Zigbee to
connect to the nodes and Ethernet, WiFi, and 3G interfaces to
connect to the Internet. This allows the Waspmote-based
WSN to be integrated into wireless 802.11 mesh networks
and hence providing a large scale WSN.
In the proposed architecture, the Waspmote sensing units are
set to form a self-organizing low power Zigbee-based
wireless mesh network. ZigBee operates in the unlicensed 2.4
GHz, 915 MHz and 868 MHz ISM bands with data rates from
20-250 Kbps. ZigBee is optimized for a low power operation
so that the devices can be operated from battery power for a
longer period of time. A Meshlium gateway equipped with a
Wi-Fi module collects and aggregate data from the base
station (BS) of a sensor network and transmits to
database/Web servers through the Internet. It can also be
configured to control WSN using a Web application. Data
aggregation is used to ignore the unwanted part of the data
and to reduce the number of redundant data, and thus only the
necessary data is forwarded to the BS. This reduces the
number of data transmissions allowing more energy
efficiency of the WSN to be achieved.
Database & Data Management System (Data Center)
This is a repository of sensors data that are sent by gateways
and processed by the Web applications. This is also known as
public or end users data center. Data are sent to and stored in
the database after being processed by a gateway. A Database
Management System (DBMS) is used to manage all data.
DBMS consists of a registry, classifier, and security services.
A registry service is used to register gateway information and
authorize users to use specific data. A classifier is also used
in DBMS to classify different types of data that could be
shared by other applications. Different security measures are
also a part of DBMS to protect sensor data from unwanted
users.
Servers
The proposed framework consists of a database server, a Web
server, alert notification service, and decision support
systems (DSS). Most of the monitoring projects require alert
notifications. Thus, the alarming component can be extracted

and reused in different projects only by changing the

parameters. Users subscribe to the alert notification service
and set the threshold value for each parameter. Sensor data is
sent to alert services at the same time it is sent to the data
center. If the sensed data value of a certain parameter exceeds
the threshold set by a subscribed user, then e-mail and SMS
messages are sent to the concerned user. Based on users
queries, Web servers collect the data from WSNs through the
gateway or the database server. Thus, the Web application
provides users with the online real time monitoring of the
sensor data and also allows the retrieval and control of
sensors remotely. This is also known as Sensor Web (SW)
that allows computers to access WSN remotely. Users can
also retrieve the previously stored (non-real time) data from
the database server as well as store the processed data on the
database server (if required).
Decision Support System
Gathered data from the WSN-based monitoring system can
be used not only to issue after-the-fact warnings but also to
predict future air quality by building an air quality model
based on analyzing historic data. Using this model, a decision
support system (DSS) can be constructed to optimize
resource utilization, maximize requirements/output, and
minimize environmental hazards. DSS allows a long term
strategy planning, promotes learning and training, generates
new evidence in support of a decision, and reveals new
approaches to solve a particular problem pertaining to the
design of monitoring systems.
WSN for Air Quality Monitoring

Waspmote
Gases Sensor

Waspmote
Gases Sensor

Waspmote
Gases Sensor

Waspmote
gateway/BS

Waspmote
with Wi-Fi
Module

Internet

Email/
SMS to
users

Alert
Notification
Service

Web
Server

Web
Application

Database/
DBMS

Expert and Decision

Support System

Industries
for decision
making on
an incident

Figure 1: Architecture of Air Quality Monitoring System

B. Routing Protocol
We introduce a simple Clustering Protocol of Air Sensor
network (CPAS) for air quality monitoring. Industrial areas,
cities major roads, indoor places such as schools, office
spaces, large housing complexes, and apartment buildings
should be monitored to detect the presence of any harmful
gases such as CH4 and CO. These might be industrial
chemical byproducts or might be due to a fire or a leakage ...
Gas sensors are deployed in the areas of interest. A simple but
effective clustering scheme is used to control a large number
of sensors that are deployed in different places to be connected
to a central station. Figure 2 illustrates the architecture of the

547

2014 International Conference on Computing, Networking and Communications, Wireless Ad Hoc and Sensor Networks
Symposium

proposed CPAS clustering scheme. Initially, the base station

(BS) randomly selects, for each cluster, a node to become a
cluster head (CH). The BS also selects the data sensing and/or
transmission interval and informs each node through the
corresponding CH. CHs select only a few gas sensors as active
based on the sensing range and the monitoring area. The rest
of the sensors in a cluster remain in sleep mode by turning
their radios off. The duty cycle of each active sensor node is
adjusted based on the type of sensors and application
requirements. Gas sensors require a certain time to heat up
before being able to reliably read gas concentration level
(known as wakeup latency). If any sensor reads data before
the wake-up latency has elapsed, the sensed data is considered
invalid. Thus, the sensors should be in duty cycle (ON mode)
for the period t on t wakeup t data, where t wakeup and

t data are the wake-up latency and the data sensing time
respectively.
Air quality does not vary greatly over a given geographical
area. Thus, not all sensors have to be active at the same time
and providing network coverage over the whole geographical
area is not a significantly important issue in the design of the
CPAS routing protocol. This fact is taken into account when
selecting the number of active gas and temperature sensors.
The CHs select the nodes with the highest energy-distance
E
vector, ED , where E and d are the residual energy of a
d
node and its distance to the CH respectively. To balance the
energy consumption, a new selection operation is performed
after a certain number of rounds which comprise a certain
number of data sensing.
After a certain time interval (determined based on the
application area and type), active gas sensors wake-up and
transmit data to CH. Once a CH collects data from all the
active member nodes it aggregates and transmits data to the
BS only if BS is within the communication range of CH. Most
existing approaches consider only a single BS or sink node.
Otherwise, a CH forms a path to BS through other CHs and
gateway nodes (for inter-cluster communications). To increase
the probability that each CH can transmit data directly to a BS,
the proposed framework considers multiple (in our
framework, only 2) BS as is illustrated in Figure 2. During
the initial network setup, a CH decides whether it can directly
transmit data to BS or whether it needs to select gateway
nodes and a data transmission path to transmit data to BS.

Waspmote
Base station

CH

Cluster 1

CH

Cluster 2

Active
gas sensor

Cluster 3
CH

CH
Inactive
gas sensor

Gas sensor
Cluster 4

Waspmote
Base station

Figure 2: CPAS Routing protocol

V. PERFORMANCE EVALUATION
This section analyzes the performance of the proposed
Clustering Protocol of Air Sensor network (CPAS) in terms of
network energy consumption and presents the simulation
results.
A. Energy Model
The CPAS protocol uses the energy model that is proposed in
[6]. If a node transmits data of size ndata to another node at
distance d
(1)
ETX ndata data ndata d 2 air
(2)
ERX ndata data
In (1) and (2), ETX and ERX represent energy consumptions

for transmitting and receiving data respectively. data and

air represent the energy spent in transmitter electronics
circuitry and RF amplifiers for propagation loss respectively.
This energy model sets the transmitting energy consumption
of a sensor node to be directly proportional to d2 where d is
the distance between the sending and the receiving nodes (a
node here might be a BS).
B.
Performance Analysis
Let us assume that N is the total number of nodes in the
network, nc is the number of clusters (and CHs) in the
network, di is the distance from the i-th CH to BS and nai is the
number of active nodes in i-th cluster. Thus, total energy
consumptions, ETX CH and E RX CH for transmitting data
from CHs to BS and receiving data from active cluster
members to CH are respectively:
nc

ETX CH n data data n data d i 2 air

i 1

(3)

nc naj

E RX CH n data data
j 1 i 1

(4)

Similarly, the energy consumption for an active node, A, for

transmitting data to its CH at distance dA is
ETXA ndata data ndata d A 2 air
(5)

548

2014 International Conference on Computing, Networking and Communications, Wireless Ad Hoc and Sensor Networks
Symposium

Thus, the total energy consumptions for transmitting data from

active nodes to CHs is
nc naj

ETX 2 n data data n data d i 2 air

j 1 i 1

(6)
Again let us assume that over a period T each node has only
one time slot of duration t units of time. Thus, the total energy
consumption of active nodes and CHs in wakeup state over T
units of time is
E wakeuptotal ETX CH E RX CH ETX 2
(7)
Also, we assume that the total number of active nodes over the
period T is

Thus, the CPAS clustering protocol is more energy efficient as

compared to the well-known protocols such as LEACH, DSC,
FTEEC (uses a sink BS).
C.
Simulation Setup and Results
We use randomly connected Unit Disk Graphs (UDGs) on an
area of 500 meters x 500 meters as a network simulation
model. Since the monitoring parameters (e.g. gas
concentration, temperature) do not frequently change in the
air, the sensing interval will be adjusted accordingly. For
instance, in day time on city roads, gas sensors are adjusted to
transmit gas and particles concentration data frequently due to
heavy traffic. Simulation parameters and their respective
values are presented in Table I.

nc

NA = n aj

Table I. Simulation Parameters and Their Values

Parameter
Value
Network area
500 m X 500 m
Number of cluster
4-8
Number of gases sensors
Max 200
Data packet size
24 Bytes
Energy consumptions for sending data packets
50 nJoule/bit
Energy consumptions in free space/air
0.01 nJoule/bit/m2
Initial node energy
2000 Joules

j 1

(8)
Thus, (N- NA) nodes will remain in idle or sleep state during
this period. If idleis the energy consumption in idle state for
one time unit, the total energy consumptions of these inactive
gases sensor nodes in idle state is

Eidle1 ( N N A ) idle T

(9)
Moreover, if the duration of a time slot for an active node is t
time units over the period T using the TDMA scheme, each
active node also remains in sleep state for about (T-t) time
units. Thus, total energy consumptions for all active nodes in
idle state is

We measure the performance of the CPAS protocol and

compare with the well-known low energy adaptive clustering
hierarchy (LEACH) routing protocol [6] in terms of network
energy consumption and number of message transmissions by
varying the number of nodes and data sensing interval.

nc naj

Eidle2 (T t ) idle
j 1 i 1

(10)

Thus, the total idle energy consumptions is

Eidletotal Eidle1 Eidle2
(11)
From (7) and (10), we calculate the total energy consumption
of the network over a period T.
Etotal E wakeuptotal Eidletotal
(12)
If EN is the initial energy of each node (in Joule), the total
network energy is
NetEnergy EN N
(13)
Thus, the network lifetime of the proposed protocol is

NetEnergy T
Netlifetim e (TimeUnit )
Etotal

Figure 3: Energy consumption varying number of nodes

(14)
In the proposed CPAS protocol both active nodes of a cluster
and CH transmit data directly to the CH and BS respectively.
Thus, active nodes do not have any receiving energy
consumption. This is because multiple base stations are
deployed in the proposed clustering scheme if a single BS is
far from the CHs of remote clusters. Thus, the CHs can
transmit data directly to BS rather than transmitting through
other active nodes and CH using multi-hop in existing
approaches. The work done in [8] finds that direct data
transmission is more energy efficient than those approaches
that use multi-hop transmissions (using Gabriel and MiniGabriel graph topologies in sensor networks).

549

Figure 4: Message transmissions varying number of nodes

2014 International Conference on Computing, Networking and Communications, Wireless Ad Hoc and Sensor Networks
Symposium

Figure 3 illustrates that the network energy consumption in the

CPAS routing protocol is much lower than that in the LEACH
routing protocol for varying number of sensor nodes. This is
because CPAS selects a set of nodes in each cluster that
remain in active state and also provides network coverage.
The rest of the sensor nodes remain in sleep mode to reduce
energy dissipation. Moreover, a CH can transmit aggregated
data to the closest BS using the concept of multiple BS
whereas in other approaches, a CH transmits data to the
distant and single BS. Similarly, Figure 4 shows that the
number of data transmissions in LEACH is much higher than
that in the proposed CPAS protocol since all member nodes of
a cluster send data to CH using TDMA scheme in LEACH.

Figure 5: Energy consumption varying data sensing interval

Moreover, this work introduces an energy efficient routing

protocol referred to as Clustering Protocol of Air Sensor
network (CPAS) for this WSN-based air quality monitoring
system. CPAS uses multiple base stations (BS) and
periodically transmits data over a short distance to a base
station.
REFERENCES
[1] K.A. Delin, and S.P. Jackson, The Sensor Web: A New Instrument
Concept, SPIE Symposium on Integrated Optics, San Jose, CA. Jan 2003.
[2] Vlgyesi, P., Ndas, A., Koutsoukos, X., and Ldeczi, . 2008, Air
Quality Monitoring with SensorMap, In Proceedings of the 7th
international Conference on information Processing in Sensor Networks, pp.
529-530, April 2008
[3] Shu-Chiung Hu, You-Chiun Wang, Chiuan-Yu Huang1, and Yu-Chee
Tseng, A Vehicular Wireless Sensor Network for CO2 Monitoring, IEEE
Sensors 2009 Conference, New Zealand
[4] North, R., Richards, M., Cohen, J., Hoose, N., Hassard, J. and Polak, J.,
A mobile environmental sensing system to manage transportation and urban
air quality, Circuits and Systems, 2008. ISCAS 2008. IEEE International
Symposium on, pp. 1994 1997, May 2008.
[5] Young Jin Jung, Yang Koo Lee, Dong Gyu Lee, Keun Ho Ryu and
Nittel, S., Air Pollution Monitoring System based on Geosensor Network,
Geoscience and Remote Sensing Symposium, 2008. IGARSS 2008. IEEE
International, Vol. 3, pp. 1370 1373, July 2008.
[6]
W.B. Heinzelman, A. P. Chandrakasan, and H. Balakrishnan. An
application-specific protocol architecture for wireless microsensor
networks.Wireless Communications, IEEE Transactions on, 1(4):660 - 670,
Oct 2002.
[7] Jelicic V., Magno M., Paci G., Brunelli D., Benini L., Design,
characterization and management of a wireless sensor network for smart gas
monitoring, Advances in Sensors and Interfaces (IWASI), 2011 4th IEEE
International Workshop on, pp.115-120, 28-29 June 2011.
[8] Karim L., Salti T.E, Nasser N.and Mahmoud Q.H., The Significant
Impact of a Set of Topologies on Wireless Sensor Networks, Journal of
Wireless Communications and Mobile Computing, Published online: 26
JAN 2012, DOI: 10.1002/wcm.1240.
[9] Kavi K. Khedo, Rajeev P. and Avinash M., A Wireless Sensor Network
Air Pollution Monitoring, International Journal on Wireless and Mobile
Networks (IJWMN), Vol. 2, No. 2, May 2010.
[10] Jen-Hao Liu, Yu-Fan Chen, Tzu-Shiang Lin, Da-Wei Lai, Tzai-Hung
Wen, Chih-Hong Sun, Jehn-Yih Juang, Joe-Air Jiang, Developed urban air
quality monitoring system based on wireless sensor networks, Sensing
Technology (ICST), 2011 Fifth International Conference on, pp.549-554,
Nov. 28 2011-Dec. 1 2011.
[11] Waspmote, libelium. http://www.libelium.com/products/waspmote/,
Accessed on March 18, 2013.
[12] http://research.microsoft.com/nec/senseWeb/
[13] http://portal.opengeospatial.org/files/?artifact_id=25562
[14] http://www.co2meter.com/?gclid=CO-Ttsrny7MCFcq7zAod6jcAMw

Figure 6: Message transmissions varying data sensing interval

Figures 5 6 demonstrate the similar performance of the
proposed CPAS in terms of network energy consumption and
number of data transmissions over different data sensing
intervals (in seconds). Both network energy consumption and
data transmission are inversely proportional to the data
sensing interval.
VI.

CONCLUSION AND FUTURE WORK

This paper introduces a Wireless Sensor Network (WSN)based air quality monitoring system using Libeliums
Waspmote and gases sensors. This system is very simple as
compared to the existing air quality monitoring systems.

550