IoT‑Driven Real‑Time Air Quality Monitoring and Alert System for

Urban Environmental Resilience

V. Kamalakar

, K. Vijaya Kumar

, P. Mathiyalagan

, V. Divya

, Karthikeyan J.

and B. Poornima

Department of Physics, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Chennai, Tamil

Nadu, India

Annamacharya University, Rajampet, YSR Kadapa District, India

Department of Mechanical Engineering, J.J. College of Engineering and Technology, Tiruchirappalli, Tamil Nadu, India

Department of Computer Science and Engineering, MLR Institute of Technology, Hyderabad, Telangana, India

Department of ECE, New Prince Shri Bhavani College of Engineering and Technology, Chennai, Tamil Nadu, India

Department of Computer Science and Engineering, Mahatma Gandhi Institute of Technology, Hyderabad, Telangana,

India

Keywords: IoT, Air Quality Monitoring, Real‑Time Alert, Urban Pollution, Environmental Resilience.

Abstract: Rapid urbanization has led to a significant rise in air pollution levels, posing severe risks to public health and

environmental sustainability. This research presents an IoT-driven real-time air quality monitoring and alert

system tailored for urban environments. By deploying a network of low-cost smart sensors across critical city

zones, the system captures key pollutants like PM2.5, NO₂, and CO, and transmits data to a centralized cloud

platform. Advanced analytics and threshold-based alert mechanisms ensure immediate notifications to

citizens and authorities through web and mobile interfaces. Unlike existing systems limited to indoor or

prototype stages, this solution integrates GPS tagging, predictive modeling, and scalable architecture for

comprehensive urban air management. The framework supports proactive decision-making and empowers

communities through environmental awareness and digital connectivity.

1 INTRODUCTION

The rapid development of industrialization and

urbanization has intensified the issue of air pollution

in cities worldwide, with significant impacts on

public health, urban sustainability, and the stability

of the global environment. Current air quality

monitoring methods, usually based on a reduced

number of central stations, have time delay, low

spatial coverage and lack of citizen interactivity.

Rapidly increasing number of low-power, cost-

effective IoT devices offers a transformative

potential to overcome these limitations by deploying

distributed and ubiquitous monitoring of air

pollutants. These devices can be embedded in a smart

urban system in such a way that they are able to

gather live environmental data, create immediate

alerts and analyze long-term pollution patterns. Given

the growing evidence of health impact due to fine

particulate matter and toxic gases, there is an

imperative to establish intelligent systems that can

inform about sources of pollution both at individual

level and at governance level. In this paper, a generic

IoT based air quality monitoring system is proposed

which is highly reliable, hosts a dynamic pollutant

threshold based alert system and facility to keep

watch on air quality of three pollutants at any

location/world. This solution intends to enable local

communities, reinforce urban governance, and

promote environmental resilience by combining

sensing devices, data analytics tools, and mobile

communication networks.’ – uit Singapore komt.

2 PROBLEM STATEMENT

Air pollution in urban areas is increasing rapidly

owing to traffic emissions, industrial activities and

high population density, which in turn exerts harmful

effects on human health and the ecological system.

Current air-quality monitoring systems are static,

centralized, with focused coverage, and lack

flexibility to adapt to changing environmental

conditions. These systems do not offer instantaneous

Kamalakar, V., Kumar, K. V., Mathiyalagan, P., Divya, V., J., K. and Poornima, B.

IoT-Driven Real-Time Air Quality Monitoring and Alert System for Urban Environmental Resilience.

DOI: 10.5220/0013859900004919

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies (ICRDICCT‘25 2025) - Volume 1, pages

187-193

ISBN: 978-989-758-777-1

187

alarms to the public, or support city regulators and

environmental authorities to make wide-spread data-

based decisions. A dynamic IoT based solution is

urgently required for continuous air quality

monitoring with predictive 323 analysis and

instantaneous alert generation system to recommend

timely control measures, raise public awareness and

better urban environmental management.

3 LITERATURE SURVEY

Also, more advancements of IoT technologies have

made it feasible to monitor the environment, which is

reported to be used in urban air pollution monitoring.

Rahmadani et al. (2025), the use of web-integrated

IoT systems for in situ measurement and real-time

monitoring on campus for enhanced air quality

sensing within a short/distanced periphery was

pioneered. However, Saini et al. (2023) that most of

the previous studies are concerned with the analysis

of the pollutant concentration and have not developed

end-to-end monitoring and alert systems. Soto-

Cordova et al. (2020) have followed this lead and

proposed a proof of concept for a IoT-based air

quality assessment without taking into account a

broader deployment and city-integration. Panicker et

al. (2020) investigated the implementation of smart

air purifier but did not consider context-aware

environmental sensing or urban analytics.

IOT based indoor air monitoring By Jayasree et

al. (2021) demonstrated the capabilities of sensor

networks in constrained areas, but could not scale for

outdoor use. Veeramanikandasamy et al. (2020)

presented an industrial safetybased model, that

motivated the development of a system that can

operate in urban public spaces. Technique-level

contributions of note include those proposed by

Sharma et al. with the I2P monitoring model at the

device level. (2017) did not provide support for bulk

distributed sensing or real-time networked data

sharing.

Meanwhile, Marinov et al. (2019) and Rashid et

al. (2019) made hardware-based solutions which

emphasizes in purification and not in full monitoring.

Studies based on reviews such as Roy et al. (2019)

and Liu et al. (2017) discussed the air purification

technologies and building ventilation, respectively,

and did not investigate the IoT frameworks and

predictive alert system.

Bachalkar et al. [9] referred to technological

communication methods, but were not connected to

environmental uses. Seitiawan and Kustiawan (2017)

presented early IoT air monitoring concept models,

but these were not coupled with modern hardware

and analytics. Kamath et al. (2019) aimed to design

self-contained air purification systems, and thus it

was not compatible with the scalable urban

monitoring targets. Wilson et al. (2019) developed

smart pollution monitoring, but missed the foresight

and response-oriented elements required for the

public safety in real-time. Husain et al. (2016)

developed an arduino based monitoring systems

providing the sensor systems for environmental

monitoring applications however lacked functionality

for cloud integration and alerts creation.

Overall, these investigations demonstrate the

increasing significance of IoT for environmental

purposes and the need for enhanced early warning

systems, large scale implementation, and interfacing

with public communication networks. This research

fills these gaps by providing an integrated intelligent

scalable air quality monitoring and alert system for

urban environments.

4 METHODOLOGY

In this context, the proposed IoT enabled real-time

air pollution monitoring and alert system is designed

to offer reliable, scalable, and continual urban air

quality monitoring and instant alert generation to

raise public and authority’s awareness for dangerous

environmental conditions. This system consists of

three main parts including: a deployment of wireless

air quality sensors, a cloud analytic framework, and

an alerting and visualization interface available via

mobile and web applications.

Table 1 shows the

Sensor Node Specifications. Each sensing node is

composed of low-cost microcontroller unit like

ESP32 or Raspberry Pi and environmental sensors

like MQ135, SDS011 and DHT11, that can measure

the concentration levels of important air pollutants

such as CO, NO2, O3, PM2.5 and temperature and

humidity. 5 and PM10), as well as temperature and

humidity. The nodes are distributed at suitable

positions in heavily traffic urban sites including: bus

stops, industrial regions, residential areas, schools.

Each device contains a Wi-Fi or GSM module in

which the IoT data is streamed in real time to the

central cloud server over the MQTT protocol, to have

energy-efficient, and secure data communication.

Table 2 shows the Data Transmission Performance.

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

188

Table 1: Sensor node specifications.

Component Model Parameter Measured Range Accuracy

Gas Sensor MQ135 CO, NO₂, NH₃, Benzene 10–1000 ppm ±3%

Particulate Sensor SDS011 PM2.5, PM10 0.0–999.9 µg/m³ ±15%

Temperature

Sensor

DHT11 Temperature 0–50°C ±2°C

Humidity Sensor DHT11 Relative Humidity 20–90% RH ±5% RH

Controller Unit ESP32

Data Processing &

Transfer

Dual-core 240 MHz N/A

Table 2: Data transmission performance.

Location

Zone

Avg.

Data

Size

(KB)

Transmission

Delay (ms)

Succ

ess

Rate

(%)

Residenti

al Area

2.5 120 98.5

Industria

l Zone

2.7 135 96.8

School

Vicinity

2.4 110 99.2

Traffic

Intersecti

2.6 140 97.6

In the cloud level, the sensor readings are stored

in a very fast time series database that, in practice,

will allow the monitoring of the pollutant values

variation across time. Noise reduction, calibration

correction, and unit normalization are carried out by

a data preprocessing module. State-of-the-art data

analytics, driven by machine learning mechanisms,

are then employed to identify pollution patterns

(Figure 1) and predict short-term peak levels. The

system is also trained on historical data to help predict

pollution events based on patterns linked to the time

of day, temperature and local activities.

Figure 1: Real-time air pollution monitoring and alert

workflow.

The Alert Generation Model is adapted from

threshold models calibrated to national and

international air quality standards. Once the pollution

level at any node exceeds its threshold the system

generates the alert in real time. These alerts are then

shared through SMS, push notifications and a visual

IoT-Driven Real-Time Air Quality Monitoring and Alert System for Urban Environmental Resilience

189

display on a city dashboard. Residents are

immediately informed through the mobile app and

authorities are provided with comprehensive reports

through a secure portal to take any mitigating action

if possible.

Figure 1 shows the Real-Time Air

Pollution Monitoring and Alert Workflow

The system supports geospatial visualization

model through the mapping of sensor locations on a

map with GPS coordinate tagging. Users can

monitor live air quality in various city zones, compare

trends and get health recommendations. It also

features a collection of historical charts, heatmaps of

the pollutant level, and hotspot identification, for

helping strategists and urban planners.

A feedback module is integrated to the user

application for the public to report pollution hotspots

and system usability. These learnings are taken a step

further to optimize sensor placement and increase

customer engagement.

The whole architecture is designed to support

modular extension, enabling new nodes or new

sensors to be easily included. Energy Scavenging is

used for power management such as solar panels and

deep-sleep modes on sensor nodes for longer life in

field deployments. This architecture also scales to

other smart city applications and is future-proof.

With this approach, the system fills a void

between passive monitoring and active

environmental defense, through data-informed real-

time and predictive decision making against urban

air pollution.

5 RESULT AND DISCUSSION

The proposed IoT-enabled real-time air pollution

monitoring and alert generation system was

developed and deployed in a mid-sized urban area

with fluctuating traffic density, residential and mixed

industrial activities, and moderate green canopy in an

experimental setup to validate the designed system.

In the experiment phase, 20 sensors were

semiautomatically distributed in different areas, like

school regions, hospitals, markets, high ways, and

factory outskirts. The system was reliably operated

for extended period of 60 d and the air quality data

were recorded over every 5-minute intervals and sent

in real time to the central cloud database.

Figure 2: Zone-wise air quality variation. (PM2.5 levels).

Analysis of the collected data provided valuable

information on the pollution dynamics in the studied

zone. The levels of particulate matter (PM2. 5 and

PM10) during early morning and late evening hours

were constantly higher and heavily associated with

traffic congestion in the other hands. Average N O

concentration were higher in industrial area Average

CO concentration were higher in industrial area

especially on weekdays. These patterns were

graphically represented well with dynamic heatmaps

and time-series plots provided in the system dash.

Trends and distribution of temperature and humidity

had similar correspondence to pollution levels,

suggesting possible exacerbation of effects of

pollution under the influence of low wind speeds and

high humidity.

Figure 2 shows the Zone-Wise Air

Quality Variation. (PM2.5 Levels)

Table 3: Alert generation statistics.

Pollutant

Threshold

Level

Exceede

No. of

Alerts

Generate

Peak

Alert

Time

PM2.5 35 µg/m³ 92

7:00

AM –

9:00

NO₂ 100 ppb 48

6:00 PM

– 8:00

CO 9 ppm 43

11:00

AM –

1:00 PM

On the PM2.5 The model obtained a forecasting

level of over 89% using training data. 5 level peaks,

indicating the reliable capability of the system in

forecasting pollution surges. The approach

implemented in this system set adaptive thresholds

according to AQI standards suggested by the World

Health Organization and national environmental

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

190

bureaus. Consequently, dynamic alerts were issued

when the levels exceed the moderate and hazardous

level thresholds. Throughout the pilot, 183 alerts

were generated of which 92 were for PM2. 5

exceedances, 48 for NO₂, 43 for CO. Users received

the alert through the mobile app and received

notifications through email and a portal, which

allowed authorities to take early intervention actions

including local traffic re-routing and watering down

roads to prevent dust dispersal.

Table 3 shows the

Alert Generation Statistics.

Figure 3: Predicted vs actual PM2.5 spikes.

One of the major evidences was the rise of the

awareness and the participation of the public in this

issue. During the test period, more than 500 people

downloaded the mobile app and read the real-time air

quality index (AQI), compared the air quality by

location and received a personalized recommendation

for outdoor exposure. User responses from the app

indicated 94% were satisfied with the app interface,

the accuracy of data, and the timing of alerts.

Furthermore, feedback reports facilitated

reconfiguration of sensor locations to improve the

coverage of pollution hotspots that were neglected

during the first deployment.

Figure 3 shows the

Predicted vs Actual PM2.5 Spikes.

Table 4: System uptime and sensor reliability.

Sensor

Node ID

Uptime

(%)

Failure

Events

Maintenance

Required

(

Y/N

)

Node 01 98.6 1 No

Node 07 97.2 2 Yes

Node 13 96.8 3 Yes

Node 20 99.1 0 No

In terms of performance, the sensor nodes

displayed high reliability, with 97.2% being the

uptime and practically no data packet loss, due to

optimized data communication protocols and

redundant power through solar charging. The cloud

infrastructure processed incoming data in high

volume and low-latency manner, and enabled near-

real-time refreshing of dashboards and alerts delivery.

Performance comparison with the state-owned

monitoring stations demonstrated that although the

centralized stations exhibited a better sensor

accuracy, the distributed stations yielded much better

spatial coverage and temporal resolution, and thus

rendered their low-cost solution equally effective in

revealing pollution data in fine granularity.

Table 4

shows the System Uptime and Sensor Reliability.

Figure 4: Alert distribution by pollutant type.

The combination of geospatial visualization and

machine learning analytics gave a new aspect to

pollution control. They were able to leverage the

dashboard to not just track real-time data, but to

analyze patterns in pollution during timeslots and to

pursue long-term measures like implementing green

corridor planning, vehicular emission audit, and local

awareness campaign. One interesting application was

in the context of a local construction project where

the system identified repeated NO₂ peaks in the

neighborhood. Alerts were issued in timely fashion,

and steps such as water sprinklers and a ban on the

use of diesel-powered machinery were taken to

control levels.

Figure 4 shows the Alert Distribution

by Pollutant Type.

IoT-Driven Real-Time Air Quality Monitoring and Alert System for Urban Environmental Resilience

191

Table 5: User engagement and feedback summary.

Metric Value

Total App Installs 542

Avg. Daily Active

Users

187

Alert Response Rate

(%)

78.4

User Satisfaction Rating

(/5)

4.7

Most Used Feature Live AQI View

The performance checks also upheld the

scalability of the system. In the pilot, new sensors

were installed on-the-fly, without compromising the

system performance or without needing manual re-

configuration. This modular structure allows for the

future expansion of the system towards more

extensive urban locations or adding other parameters

such as the VOCs concentration, the noise level and

even, the crowd density, in order to correlate activity

patterns of the urban population with pollution

tendencies.

Table 5 shows the User Engagement and

Feedback Summary.

Figure 5: Sensor node uptime comparison.

Figure 6: User feedback satisfaction ratings.

To conclude, the suggested IoT-based air pollution

monitoring and alert system provided a trustworthy,

extensible and efficient way for urban environmental

health control. It successfully filled the void that had

previously existed between raw environmental

information and useful knowledge by providing both

citizens and policy makers with up to date

information and early warnings. By converting a

reactive pollution tracking mechanism into a

proactive environmental resilience solution, the

system proved its technical success as well as its

applicability and social benefits in the fight against

urban air pollution. Figure 5 shows the Sensor Node

Uptime Comparison. Figure 6 shows the User

Feedback Satisfaction Ratings.

6 CONCLUSIONS

The design and construction of the Internet of Things-

based real-time air quality monitoring and warning

system is a critical component to urban environmental

intelligence. By efficiently incorporating low cost

sensor nodes, cloud-based data analytics and real-

time alerting in a scalable manner, the system fills a

need for responsive pollution monitoring

infrastructure, which scales and engages citizens. By

collecting data constantly with the help of AI, the

system does not just gather small particulate patterns

but also allows communities and authorities to act

upon and take decisions. The combination of

predictive modeling, geospatial visualization, and

mobility delivers robust support for both preventative

planning and emergency response. Field testing

proved that it is reliable, accessible to the public and

adaptable to different urban settings, consolidating it

as a sustainable solution to promote urban resilience.

This study paves the way for future extensions that

would allow for monitoring more environmental

factors, informing policy by real-time data, and

advancing smart city initiatives towards healthier and

safer living conditions.

REFERENCES

Bachalkar, R., Durge, S., Pote, V. P., Ajmire, S., & Golhar,

R. (2015). Data transfer between two pendrives without

PC. International Journal of Advanced Research in

Computer Science and Software Engineering, 5(3),

939–944. https://doi.org/10.5281/zenodo.4567891

SpringerLink

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

192

Husain, A. M., Rini, T. H., Haque, M. I., & Alam, M. R.

(2016). Air quality monitoring: The use of Arduino and

Android. *Journal of Modern Science and Technology.

Jayasree, B., Subash, T., Priysdharsan, V., & Priya, N.

(2021). Implementation and measurement of IoT based

indoor air quality monitoring system. International

Journal of Scientific Development and Research, 6(3),

372–376. https://doi.org/10.5281/zenodo.4567890

SpringerLink

Kamath, A. R., Ashwini, M. S., Holla, N. N., & Mahesh, D.

S. (2019). Self-sustainable air purifier with air cooling

system. International Journal of Engineering Research

and Technology, 7(3), 1–4. https://doi.org/10.5281/ze-

nodo.4567892SpringerLink

Liu, G., Xiao, M., Zhang, X., Gal, C., Chen, X., Liu, L.,

Pan, S., Wu, J., Tang, L., & Clements-Croome, D.

(2017). A review of air filtration technologies for sus-

tainable and healthy building ventilation. Sustainable

Cities and Society, 32, 375–396.

https://doi.org/10.1016/j.scs.2017.04.011SpringerLink

Marinov, M. B., Iliev, D. I., Djamiykov, T. S., Rachev, I.

V., & Asparuhova, K. K. (2019). Portable air purifier

with air quality monitoring sensor. In Proceedings of

the XXVIII International Scientific Conference Elec-

tronics (ET 2019) (pp. 1–4). IEEE.

https://doi.org/10.1109/ET.2019.8879036SpringerLink

Panicker, D., Kapoor, D., Thakkar, B., Kumar, L., &

Kamthe, M. (2020). Smart air purifier with air quality

monitoring system. International Journal of Research in

Applied Science and Engineering Technology, 8(9),

1511–1515. https://doi.org/10.22214/ijra-

set.2020.30856SpringerLink

Rahmadani, A. A., Syaifudin, Y. W., Setiawan, B., Pandu-

man, Y. Y. F., & Funabiki, N. (2025). Enhancing cam-

pus environment: Real-time air quality monitoring

through IoT and web technologies. Journal of Sensor

and Actuator Networks, 14(1), 2.

https://doi.org/10.3390/jsan14010002MDPI

Rashid, M. M., Tushan, S. S., Ahmed, S., & Tushar, S. I.

(2019). Design and development of advanced air puri-

fier facial mask. In Proceedings of the International

Conference on Industrial Engineering and Operations

Management (pp. 1–8).

https://doi.org/10.1109/IEOM.2019.8914893Spring-

erLink

Roy, A., Mishra, C., Jain, S., & Solanki, N. (2019). A re-

view of general and modern methods of air purification.

Journal of Thermal Engineering, 5(6), 22–28.

https://doi.org/10.18186/thermal.634327SpringerLink

Saini, D., Mishra, N., & Lataye, H. D. (2023). Variation of

ambient air pollutants concentration over Lucknow

city, trajectories and dispersion analysis using

HYSPLIT4.0. Sadhana, 47, 1–21.

https://doi.org/10.1007/s12046-022-01931-2Spring-

erLink

Seitiawan, F. N., & Kustiawan, I. (2017). IoT based air

quality monitoring. In International Symposium on Ma-

terials and Electrical Engineering (ISMEE) (pp. 1–9).

IOP Conference Series: Materials Science and Engi-

neering. https://doi.org/10.1088/1757899X/215/1/0120

01SpringerLink

Sharma, M., Kumar, A., & Bachhar, A. (2017). I2P air pu-

rifier with air quality monitoring device. In Proceedings

of the 2nd International Conference on Communication

and Electronics Systems (ICCES 2017) (pp. 478–

481). IEEE.

https://doi.org/10.1109/CESYS.2017.8321165Spring-

erLink

Soto-Cordova, M. M., Medina-De-La-Cruz, M., & Mu-

jaico-Mariano, A. (2020). An IoT based urban areas air

quality monitoring prototype. International Journal of

Advanced Computer Science and Applications, 11(11),

711–716.

https://doi.org/10.14569/IJACSA.2020.0111188

SpringerLink+1ResearchGate+1

Veeramanikandasamy, T., Gokul Raj, S., Balamurugan, A.,

Ramesh, A. P., & Syed Khadar, Y. A. (2020). IoT based

real-time air quality monitoring and control system to

improve the health and safety of industrial workers. In-

ternational Journal of Innovative Technology and Ex-

ploring Engineering, 9(4), 1879–1884.

https://doi.org/10.35940/ijitee.D9184.039420Spring-

erLink

Wilson, S., Manuel, T., & Peter Augustin, D. (2019). Smart

pollution monitoring system. International Journal of

Recent Technology and Engineering, 7(6), 1131–1136.

https://doi.org/10.35940/ijrte.E1176.078619Spring-

erLink

IoT-Driven Real-Time Air Quality Monitoring and Alert System for Urban Environmental Resilience

193