Proceedings of the Second International Conference on Inventive Research in Computing Applications (ICIRCA-2020)

IEEE Xplore Part Number: CFP20N67-ART; ISBN: 978-1-7281-5374-2

Real Time Analysis of Air Pollution Prediction using IoT

Niranjan.D.K [Link]
Department of Computer Science and Engineering, Department of Computer Science and Engineering ,
Amrita School of Engineering, Bangalore, Amrita School of Engineering, Bengaluru
Amrita Vishwa Vidyapeetham,India Amrita Vishwa Vidyapeetham, India
niranjandk6@[Link] m n_rakesh@[Link]

reduces the allergies of the person. Air pollutants is taken into

Abstract-Reducing the pollution concentration level is currently consideration as the main environmental chance factor in the
the most important in environment. Pollution are in several occurrence and growth of a few diseases consisting of asthma,
forms nearly each natural thing is currently suffering from the lung cancer, ventricular hypertrophy, mental complications,
term pollution. At present earth is currently in danger level autism, and low beginning weight.
which consists of land, water, air. Nowadays everyone has The system is designed for this situation so that the person
changed their lifestyle but changes as not made environment any
can get fresh air, in our system the polluted air is taken and by
changes because it has exceeded the air pollution. Therefore
there should be an innovative to reduce the concentration of air some purify ing techniques, the fresh air produced. Our
pollution. In this method the air is being monitored and purified system is portable, so that anyone can carry to anywhere, it is
by passing air into many purification filter stages depending on the battery operated system which consumes less power and
the air concentration levels. This method consumes less power does more effective work. The device can use in car or in
and it is portable device. home or in office premises. It is easy and low weight to carry.
The system is designed in such a way that the air purifies
Keywords - Air purification automatically depending upon the environmental condition.
The air is purified based on the condition of the region. The
I. INTRODUCTION condition are the regions of the environment, the polluted air
In our day to day life it is noticed that air is contaminated, and is monitored and purifies the air based on the system prev ious
our health risk has been increasing day to day. The air is updated data value. Based on the air pollution many research
contaminated by several factors which are all by man-made works has already been done, some of them are by using the
substances. There are t wo categories those are indoor and plants leaf and roots, and some of them done by using water
outdoor elements, those are from our daily commut ing for purification. These does not help much and not able to
vehicles, factories, waste burning, construction work, and the carry with us. As the proposed system is designed for low
domestic kitchen. These are the major factors that the air in power, the cost of the device is also less so any one can afford
the atmosphere is polluted; to reduce the polluted air is too it. The reason that motivated me to do this work is the major
tough, because all are important in our daily lifestyle. But our health effect to the human and unbalancing in the
health is also the most important consideration in our daily environment. In [1], th is paper air pollution monitoring and
life. By air pollution many health issues are faced, they are sensor calibrat ion is done Based on sensor they have done
difficulty in breathing and card iovascular dis eases, calibrat ion for accuracy. They surveyed that new air sensor
neuropsychiatric co mp licat ions, the eyes irritation, pores and consumes more power, and sensing accuracy is less in crowed
skin diseases, and long-term persistent diseases which include area. In [7], this paper, the air purificat ion system is done on
cancer. These are effected by the air polluted in at mosphere, the streets and road side air pollution, they have purified the
the most common contaminates in the atmosphere are part icle air through drain water, the draw back in this paper is, and it
pollution, ground-stage ozone, carbon monoxide, sulfur consists of millions of bacteria and needs many filters to
oxides, nit rogen o xides, and lead. These are the six purify the water. In [8], this paper, it is a IoT device for
contaminates which has been decreased human life. The checking the quality of air, and the purifying the system is
World Health Organizat ion (W.H.O) has been declared that done by using indoor plants for purifying the polluted air. The
they are very co mmon and harmful to the hu man health [3]. drawback is that it requires sunlight and water; it is limited
More air pollut ion are damaging the ozone layer and the for only indoor. Our system takes the data based on the
people are facing more cos mic light wh ich is also more regions of the environment, by day to day and the data is
dangerous to the health and to the environment. To control stored in the data base. The polluted air is taking hu man
the pollution fro m vehicles it is required to switch to the health to high risk, as the factories are producing more
electric vehicles so that only 2-3% of pollution is reduced and pollution and the construction work are also produce more
the rest remains same, to control the rema in ing is much dust particles. The dust particles in the air are of 1 to 100 µm
tougher, but it can be reduced a little bit and purify the air for in diameter of the size, the thickness of the particles is less
our health purpose. By purify ing the air is not only the than the size of our hair. These air part icles affecting the
solution but at present situation it’s the best idea, so that the breathing and causing damaging to lungs, some person are
life rate can be increased and start breathing the fresh air and facing allergies on the skin or on the eye due to impurit ies in

978-1-7281-5374-2/20/$31.00 ©2020 IEEE 904

Authorized licensed use limited to: Cornell University Library. Downloaded on September 06,2020 at 09:59:42 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Second International Conference on Inventive Research in Computing Applications (ICIRCA-2020)
IEEE Xplore Part Number: CFP20N67-ART; ISBN: 978-1-7281-5374-2

the air and the particulate matter in the air, these are the major
problems where the most metropolitan cit ies persons are
facing, the high level of carbon mono xide is emitting fro m Polluted Air Anti-Bacter
the old and heavy vehicles. This product can be used in the Air Purifying ial Filters
car so that air is purified, and will get fresh air. The IN Filters
metropolitan city person spending most of the time in
travelling to home or office and facing the severe health
problems. As the study fro m the W.H.O says that most
problems are facing by the city person and reporting most Ionizer
diseases caused by the construction work and the vehicle
emitting pollution. The proposed system has used filters to
purify the particu late matter. System data can also be Purified
monitored for the monitoring the changes in the environment Air Data System Analyse
and take the necessary action. So me air purifiers are already And
in market, but those are not carrying able and the cost is more, Data Base at the
and consumes more power. Those systems are not automatic base station
where it just purifies the air. The automatic means it depends Purified Air
on the environment range and the data stored depending on it passed out Air Pass
will work and updates the data to the web server. Data

This paper is structured as follows. In Section II, the proposed

system is explained. Section III consists of hardware
description. The Section IV describes the setup and the result Purifying Air purifier Air Pump
of practical experiments. Finally Sect ion V concludes and Stage Chamber by DC fan
future work of the paper.
Figure 1. Block Diagram of Proposed system
II. PROPOSED SYSTEM
The air quality is measured, by taking polluted air and the
The main objective of this work is to prevent fro m the details of the polluted air is recorded of every minute of the
non-communicable lung diseases, and make our surroundings cycle. The purificat ion stage is done based on three levels of
into fresh air. Our system is a real time analysis of the air purifying, those are: low stage, mediu m stage, and high stage
pollution, and purifies the air based on the real time and the purifying. The serial co mmun ication of I2C is been used for
data stored from the real t ime analysis. The proposed system communicat ing with sensor to the raspberry pi. The sensor
has used Raspberry Pi 3B+ module for real time analysis, collects the data and stores in the memory of the raspberry pi;
storing the data base and purifying the air. Three DC 12V the stored data are collected by the sensor for real time
fans are been used for the intake of the polluted air. MQ 135 analysis and depends on environment. A DC has been used for
and MQ 02 sensors are been used for collecting the air data converting the analog signal to the digital signal of the sensor.
and monitoring the contamination in the air. The system NGINX is called Engine X is been used for uploading the
purifies the air and g ives the fresh air, for purificat ion of air a sensor data to the web server for mon itoring the data based on
5-stage filter is used. the date and time, even voltage consumed by the sensor.
SQlite of version 3 is been used for record the data and store
The block d iagram o f the proposed system is shown in the the data value of the sensor, the system is designed to record
figure 1, where polluted air is inlet by the inleter and the the data for every one minute of interval. If the air quality is
purifying stages are been done. The anti-bacterial and ionizer fresh then the device just records the data but it will not run
filters are important where it removes the bacteria in the air the purification part, e.g. Village side. If the air quality is at
by anti-bacterial filter and adding the negative ions to the the border line to the fresh air then the system runs for few
fresh air, the air purifier chamber consists of purifying stages minutes and turns off, when the ever the air quality is varied
where fine dust filters, cold catalyst, activated carbon, HEPA, then the system turns ON. If the quality of air is moderate
Sterilization cotton filters. Then the air is passed out, the inlet then the system runs continuously on activating two fans for
air is monitored and stored in data base, and output data is purifying which will pass through all stages of filters. If the
also monitored. The system analyses the data and does the air quality risk is high then the system runs all the three fans
action based on the environment, and also sets the speed of and purifies the air at high speed.
the fan of purifying the air.

978-1-7281-5374-2/20/$31.00 ©2020 IEEE 905

Authorized licensed use limited to: Cornell University Library. Downloaded on September 06,2020 at 09:59:42 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Second International Conference on Inventive Research in Computing Applications (ICIRCA-2020)
IEEE Xplore Part Number: CFP20N67-ART; ISBN: 978-1-7281-5374-2

A
START
NO

Run the system

Initialize the region continuously at high speed

Consider the data set of the

regions Data Purify the air by the filtering
Base at the stages
the base
station
Initialize the sensor Fresh air is produced
value of present situation

Data value of the fresh air B

B

If the air YES Stop

quality is Read the values and
fresh stop purifying air Figure 2. Flow chart of the proposed system

The system will also notify that the air is safe or not safe,
when the air is contaminated by the poison gas in the air. The
NO proposed system has collected more than 1000 data values to
analyse the real time system, the system collects the data
based on the area or region of the device is activated. The
system collects the data of all regions, and activates when the
system activates the different regions then the data set will
If the air Run the device also change to the particular regions. The data set will change
quality is YES
for some-time and depending on the environment after one month at present.
less to the turn off till air The data are stored in the server of the base station,
borderline quality is varied depending on the situation the data are been uploaded. The
background process of the system is more than the
foreground process. The flow of the system is shown in the
above figure 2. The flow chart tells about the process of the
air purification and system working process. The system can
NO be turned ON continuously. It is like mob ile device so no
need of the power every time . The system is designed for
minimal area of covering where the person can get fresh air.

III. HARDWARE DESCRIPTION

If the air Runs the system
YES
quality is continuously with The proposed system has used Raspberry PI 3B+ for doing all
moderate running 2 fans the required process at the less power so that the system can
NO be configured perfect ly, where the raspberry pi has the built
in Wi-Fi so that the data values can be directly uploaded to
the database. Raspberry pi is of 1.4GHz 64-bit quad-core
processor, dual-band wireless LAN, Bluetooth 4.2/ BLE,
faster Ethernet, and Power-over-Ethernet support (with
separate PoE HAT), so this device is used for our work.
A

978-1-7281-5374-2/20/$31.00 ©2020 IEEE 906

Authorized licensed use limited to: Cornell University Library. Downloaded on September 06,2020 at 09:59:42 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Second International Conference on Inventive Research in Computing Applications (ICIRCA-2020)
IEEE Xplore Part Number: CFP20N67-ART; ISBN: 978-1-7281-5374-2

communicat ion. Figure 6 shows the ADS1015 ADC, for

microcontro ller, it is o f 12-b it precision at 3300
samp les/second over serial co mmun icat ion . It is of
higher-p recis ion A DC without an analog -to-d ig ital
converter. Th e ch ip can be con figured as 4 sing le -ended
input channels, or t wo d ifferent ial chann els . To the fu ll
range of th is dev ice is up to x16 p rog rammab le gain
amp lifier, wh ich helps to imp rove s maller s ignal or
different ial s ignals. It is the low po wered dev ice wh ich runs
at 2V to 5V. Where maxi mu m range of s igna l is ach ieved
when it attains at the maximu m of 5V.

The ch ip is s mall in size so it co mes on a b reakout board

with ferrites to keep the A VDD and A GND quiet.
Figure 3. Raspberry Pi 3B+ module Interfacing is done v ia serial co mmun icat ion where in
raspberry p i the I2C should be enab led fo r co mmunicat ion.
The proposed system has used gas sensor to detect the polluted It h as four opt ions where th e address can be changed so that
air and purify the air, sensors are used for detecting are it can connect up to 4 A DS1015 fo r 16 s ingle ended inputs
MQ-135 and MQ-02 sensors. Figure 4 shows the MQ135 to an I2C bus.
sensor and its pin details. M Q-135 Detects or Measures NH3 ,
NOx, alcohol, benzene, s moke, CO2 , etc. It has a wide
detecting scope, stable, fast response and high sensitivity. It is
of longer durability, and consumes only 5V. It can be used
both analog and digital depending on the work; the d igital
output consists of TTL logic

Figure 6. I2C ADS1015

A 12V DC normal fan is used for inleting the polluted air,

where it should have constant 12V supply to work perfectly.
The proposed work has used 5-stage purification filters. The
purification filters are pre-filter or fine dust filter, cold catalyst
filter, activated carbon, HEPA, Sterilization cotton, Negative
Figure 4. MQ-135 Sensor and the pin details Ion filters. Figure 7 shows the filters used in our proposed
work. Prefilter is used so that it purifies the fine dust particles
fro m the construction work or the vehicles, carbon filter for
MQ-02 sensor is also used in our work for detecting other type
purifying like dust, lint, mo ld spores, pet dander and pollen.
of gases. Figure 5 shows the MQ135 sensor and its pin details.
HEPA filter used for purifying harmful s mokes and the
It is to Measure or detect LPG, Alcohol, Propane, hydrogen, dangerous gases.
carbon monoxide and methane. It is stable and long life, and
consumes only 5V. It can be used both analog and digital
depending on the work; the dig ital output consists of TTL
logic.

Figure 7. Air purification filter stages

Figure 5. MQ-02 Sensor and the pin details

The proposed system has used 12bit I2C ADC, ADS1015 is

the device used for the converting and fo r serial

978-1-7281-5374-2/20/$31.00 ©2020 IEEE 907

Authorized licensed use limited to: Cornell University Library. Downloaded on September 06,2020 at 09:59:42 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Second International Conference on Inventive Research in Computing Applications (ICIRCA-2020)
IEEE Xplore Part Number: CFP20N67-ART; ISBN: 978-1-7281-5374-2

IV. RESULT ANALYSIS

I2C for serial Sensors
The proposed system is designed and operated in few reg ions communication
where nearly 1000 sets of data are collected for best
efficiency at the different regions. The purification of the
system will operate based on the present situation. As the data
varies it changes the data set range and applies the process
which has been done at the background of the system. The
data set is collected and store in the data base where it
collects the data for every one minute of the time cycle. The
system is well operated in the smaller areas where the
effectiveness of the system is well purified. The proposed
system can reduce our health risk fro m the affected air
pollutants. The data are stored in the SQlite, on the day to day
wise. The figure 8 shows the data recorded and stored in
SQlite.
Raspberry
Pi 3B+

Top View

Figure 9. Proposed system design (Top is front view, Middle

is rear view, and the bottom one is the top view of the system)

The polluted air is taken at the rear side of the system, and
produce fresh air fro m front side. The figure 10 show the
simu lated wave form of the air quality in the web server
Figure 8. Data is recorded base on the time set on day to day (engin x) based on the metro cities calibration. The air is safe
wise or not safe is monitored as shown in the figure 11. It is
monitored by the poisonous gas sprayed around the area for
The proposed system as shown in figure 9, where it operates best result.
on the required amount of polluted air, and does the effective
work.

MQ-02 MQ-135
Sensor Sensor

Front View Figure 10. Air quality waveform in webserver

Air Inlet
fans of
12V

Rear View Figure 11. Air is Safe or not safe notification

978-1-7281-5374-2/20/$31.00 ©2020 IEEE 908

Authorized licensed use limited to: Cornell University Library. Downloaded on September 06,2020 at 09:59:42 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Second International Conference on Inventive Research in Computing Applications (ICIRCA-2020)
IEEE Xplore Part Number: CFP20N67-ART; ISBN: 978-1-7281-5374-2

The application of the proposed system can be used in the [8] Sanjay Shitole, Devika Nair, Nidhi Pandey and Heena Suhagiya, “ Internet
of things based indoor air quality improving system” 3rd International
home, in car or transportation, for purificat ion of air. It can be Conference for Convergence in Technology (I2CT) 2018.
used in so me work places also. The device is limited for only [9] Jinhee Lee and Hyunkyung Kang “ The Effect of Improving
small area o f coverage, because it is the battery operated and Indoor Air Quality using some C3 Plants and CAM Plants” Indian
portable. Journal of Science and T echnology, Vol 8(26) 2015.
[11] [Link] Raju, [Link] and B Sangeeth Kumar, “ Pollution
Monitoring System using Wireless Sensor Network in Visakhapatnam”,
V. CONCLUSION AND FUTURE WORK International Journal of Engineering Trends and T echnology (IJETT) -
Volume4Issue4- April 2013.
[12] Dr. N. Rakesh, “Performance analysis of Anomaly Detection of
Conclusion Different IoT Datasets using Cloud Micro Services”, in International
Conference on Inventive Computation Technologies (ICICT 2016)
[13] Abul Bashar, “Review on sustainable green Internet of Things and its
The real time analysis of the air purification is done based on applications”, IRO Journal Sustainable Wireless System, 1, No. 4, 2020:
the present air quality and the data set stored in the system, 256-264.
depending on the variation of the air quality the system [14] Duraipandian.M and [Link], “ Cloud based Internet of things
for smart connected objects” , Journal of ISMAC 1, No: 02 (2019):
generates purified air to the person, the data recorded will 111-119.
help for mon itoring the environment condit ion. The system is [15] [Link] and [Link] Mary, “IoT based air pollution
worked effectively in the smaller areas. Most process of the monitoring and control system”,International Conference on Inventive
system is done at the background of the system; python Research in Computing Applications (ICIRCA 2018) Jan 2018.
language is used for processing the system and SQlite for
recording the data set. The engin x is used to check the results
of the uploaded data. The air is purified based on the region,
previous air data, and the present air quality. It is less power
consumable and it is the battery operated, and easy to carry
any-where for our health purpose. Thus our device helps to
reduce the health effects that are caused by the polluted air.

Future Work

In our work there are some limitations, so the work is

expanded to the next level so, that it is still more p recised.
Solar panels will be added so that there will be no need to
charge every-time. Application for monitoring the system and
remotely turning ON and OFF the device is also being in
progress in our work, and also planning to add GPS location
to the device such that the system can be done the centralized
system.

REFERENCES

[1] Balz Maag, Zimu Zhou, Lothar Thiele “A Survey on Sensor Calibration
in Air Pollution Monitoring Deployments”, IEEE Internet of Things
Journal, Volume 5, Dec 2018.
[2] Camilo Cerro “Developing solutions for dealing with water and food
scarcity”, Advances in Science and Engineering Technology
International Conferences (ASET) Dubai Feb 2018.
[3] Vikram Reddy, Deep Air: Forecasting Air Pollution in Beijing, China
(2017)
[4] Curtis, L, Rea, W, Smith-Willis P, Fenyves, Pan Y. Adverse health
effects of outdoor air pollutants. Environ. Int. 2006, 32, 815–830.
[5] Hastie, T., T ibshirani, R., Friedman, J. H. The elements of statistical
learning data mining, inference, and prediction. (pp. 417-458). New York,
NY: Springer 2016.
[6] Rohde, R. A., Muller, R. A. Air Pollution in China: Mapping of
Concentrations and Sources. Plos One, 10(8).
doi:10.1371/[Link].0135749, 2015.
[7] [Link] Subrahmanyam and Avanish Gautam Singh, “Air Purification
System for Street Level Air Pollution and Roadside Air Pollution”2018
International Conference on Computing, Power and Communication
T echnologies (GUCON)

978-1-7281-5374-2/20/$31.00 ©2020 IEEE 909

Authorized licensed use limited to: Cornell University Library. Downloaded on September 06,2020 at 09:59:42 UTC from IEEE Xplore. Restrictions apply.