What question did this study set out to answer?

To develop a device that monitors, identifies, and controls urban noise pollution using IoT and AI technologies.

April 1, 2026Open Access

Iot, AI, and Ml Based Noise Source Identification, Monitoring and Controlling Device With Alert Mechanism

Key Points

To develop a device that monitors, identifies, and controls urban noise pollution using IoT and AI technologies.
Built an integrated system using sensors for continuous noise monitoring.
Utilized machine learning models for noise source identification through acoustic signals.
Implemented an alert mechanism to notify users of noise regulation breaches.
Incorporated edge computing for effective data processing and real-time responses.
Achieved 89% accuracy in classifying noise sources within 3 seconds.
Successfully reduced noise level breaches by 34% through automated control measures.
Improved urban living conditions by effectively managing noise pollution.

Abstract

Recently, more and more issues associated with urban noise pollution have been addressed. That is why recent urban noise pollution appears as key problem causing millions worldwide to sleep poorly, develop cardiovascular issues, and live a welfare lower-quality life. Conventional means of monitoring sound are oriented towards inspection of noise and do not have sound analysis intelligence to act on its measure. This article discusses how an artificial intelligence and machine learning IoT based noise source isolation system works which includes continuous monitoring, and the necessary automation mechanism of control. This solution assumes the integration of sensors on one side, while state-of-the art edge computing approach incorporating trained machine learning models works on the other hand. After noise sources have been localized, the systems then go on to hold the noise levels within the regulators limits and send out alerts whenever the regulations have not been adhered to within the required duration. In this context, Acoustic signals are passed through a Convolutional Neural Network in a forward model. This network is able to discern different sources of noise such as human activity, traffic, construction, and equipment, with very close accuracy. Furthermore, data consolidation, predictive analysis, and user alerting are executed through IoT cloud platform delivered through mobile and web dashboards. In addition, we have successfully utilized the led noise infringement system in the environmental noise management system which is based on non-directional loudspeakers for premature noise control across urban, industrial and residential sectors. Noise is discouraged by the system once it notices the existence of a breach of light noise and those commitment is really a good one since 89 percent of all the noise sources present in the System will be classified under 3 seconds on average when alert is to be sent for breach of containment rules and then it will go 34% more times further in automatically reducing the breach by the defaulter. That is all, as usage of noise control solution for the smart city adhering to the requirements of noise Proposed Regulation is concerned.

Bookmark

View Full Paper