INTRODUCTION

Sound pollution has emerged as a major problem in the past few decades and its implications are very detrimental. The recent industrialization and the rate of increase in number of motor vehicles has mainly contributed to the cause. The first step of countering sound pollution is to monitor the cause of it. As it may be seen that a lot of research has been going on in the field of sound pollution indexing, considerably low attention has been paid to the origin of sound. To monitor the origin and intensity of sound we have designed a system which can efficiently monitor the intensity of sound and also determine the origin of sound using machine learning algorithms.

PROPOSED SOLUTION

To solve the problem we will design a device which will collect the data, analyse it and display the result in a mobile application. Our device will be attached to each of the street light poles in a particular area. We will collect the data using ADMP401 MEMS microphone sensor. Using esp8266 wifi module data will be uploaded to “ThingSpeak”, an open source IoT platform that enables us to collect and store data in cloud. That data will be analysed in Matlab and from that intensity of sound can be found. Sound library from Google’s database will be accessed and fed to Matlab and using machine learning techniques the source of sound will be determined. We will develop a mobile application, where these data will be displayed. In the application, when a particular place and a date is entered, a plot of intensity of sound versus time will be shown from where the pollution level can be estimated. To plot the sound intensity at a particular time, we will take the average of the data obtained from all the sensors at that place. There will be an option where we can enter multiple dates to compare such plots of a particular place. An option will also be provided to show the source of the pollution. The UI will be designed in such a way that the day with most and least pollution in the last seven days will also be shown.

TECHNICAL DETAILS-

Input voltage to NodeMcu=3.3 volt.

Max. Input current of NodeMcu =250 mA

Input voltage to ADMP401 MEMS =3.3 volt.

Max. Input current of ADMP401 MEMS =350 µA

Total power drawn= 3.3x(350x10^6+250x10^3) Watt

=0.826 watt

Since the power drawn by the device is very small, it can be powered from the same source as for the street light without affecting the backup time of the light.

COST ANALYSIS

The cost of the device includes NodeMcu, ADMP401 MEMS microphone sensor, casing & wiring and server cost. Since the server cost is dependent on the scale of use, we will only look into the hardware cost of the device.

NodeMcu- $5.5

ADMP401 MEMS- $10.95

Casing & Wiring- $3

Therefore, total hardware cost is approximately- $20

APPLICATION AND SCALABILITY

As our device will be embedded within the street light poles, the device will be applicable in those areas where the number of street light poles are more frequent. Nowadays with the increase in the number of urban and developed cities the street light poles seem to be deployed everywhere so our device can be used in those areas. Now coming to the point of scalability: as our device is cost effective and also draws much less power, this makes the device easily employable. With the use of IoT, the device is always connected and employable in any part of the world.

Inspiration

Sound Pollution has very devastating effects specially in urban areas due to rapid increase in the number of vehicles and industrialization. It has a very harmful impact on human health. It is known to cause hypertension, cardiovascular issues, sleeping disorders and it also has a great impact on wildlife. Such parameters make sound pollution a very important issue to be addressed. So, we developed a sensor network strategy for monitoring sound pollution generated at a particular place and identify its source of origin so that necessary actions can be taken by the authorities to reduce the amount of sound pollution by checking the sources creating it.

Voting

Voting is closed!